IEJ.01.2009

Effectiveness of ozone againstendodontopathogenic microorganisms in a rootcanal biofilm model

K. C. Huth1, M. Quirling1,2, S. Maier1, K. Kamereck3, M. AlKhayer1, E. Paschos4, U. Welsch5,T. Miethke3, K. Brand6 & R. Hickel1

1Department of Restorative Dentistry & Periodontology, Ludwig-Maximilians University, Munich; 2Institute of Clinical Chemistry &

Pathobiochemistry, Klinikum rechts der Isar, Technische Universitat Munchen, Munich; 3Institute of Medical Microbiology,

Immunology and Hygiene, Technische Universitat Munchen, Munich; Departments of 4Orthodontics and 5Anatomy, Ludwig-

Maximilians University, Munich; and 6Institute of Clinical Chemistry, Medizinische Hochschule Hannover, Hannover, Germany

Abstract

Huth KC, Quirling M, Maier S, Kamereck K, AlKhayer M,

Paschos E, Welsch U, Miethke T, Brand K, Hickel R.

Effectiveness of ozone against endodontopathogenic microor-

ganisms in a root canal biofilm model. International Endodontic

Journal, 42, 3–13, 2009.

Aim To assess the antimicrobial efficacy of aqueous

(1.25–20 lg mL)1) and gaseous ozone (1–53 g m)3)

as an alternative antiseptic against endodontic patho-

gens in suspension and a biofilm model.

Methodology Enterococcus faecalis, Candida albicans,

Peptostreptococcus micros and Pseudomonas aeruginosa

were grown in planctonic culture or in mono-species

biofilms in root canals for 3 weeks. Cultures were

exposed to ozone, sodium hypochlorite (NaOCl; 5.25%,

2.25%), chlorhexidine digluconate (CHX; 2%), hydro-

gen peroxide (H2O2; 3%) and phosphate buffered saline

(control) for 1 min and the remaining colony forming

units counted. Ozone gas was applied to the biofilms in

two experimental settings, resembling canal areas

either difficult (setting 1) or easy (setting 2) to reach.

Time-course experiments up to 10 min were included.

To compare the tested samples, data were analysed by

one-way anova.

Results Concentrations of gaseous ozone down to

1 g m)3 almost and aqueous ozone down to 5 lg mL)1

completely eliminated the suspended microorganisms

as did NaOCl and CHX. Hydrogen peroxide and lower

aqueous ozone concentrations were less effective.

Aqueous and gaseous ozone were dose- and

strain-dependently effective against the biofilm

microorganisms. Total elimination was achieved by

high-concentrated ozone gas (setting 2) and by NaOCl

after 1 min or a lower gas concentration (4 g m)3)

after at least 2.5 min. High-concentrated aqueous

ozone (20 lg mL)1) and CHX almost completely

eliminated the biofilm cells, whilst H2O2 was less

effective.

Conclusion High-concentrated gaseous and aqueous

ozone was dose-, strain- and time-dependently effective

against the tested microorganisms in suspension and

the biofilm test model.

Keywords: antimicrobials, biofilm, endodontics,

microbiology, ozone, root canal.

Received 18 April 2007; accepted 1 July 2008

Introduction

The successful treatment of an infected root canal,

especially those with persistent apical periodontitis

remains a clinical challenge (Nair 2006). The main

aim of endodontic treatment is to eradicate or substan-

tially reduce the microbial load in the root canal

Correspondence: Dr Karin Christine Huth, Department

of Restorative Dentistry, Periodontology & Pedodontics,

Dental School, Ludwig-Maximilians University, Goethestr.

70, 80336 Munich, Germany (Tel.: +49 89 5160 9411;

fax: +49 89 5160 9302;

e-mail: [email protected]).

doi:10.1111/j.1365-2591.2008.01460.x

ª 2008 International Endodontic Journal International Endodontic Journal 42, 3–13, 2009 3

system, which is conventionally achieved by chemo-

mechanical instrumentation followed by canal filling to

prevent recolonization (Nair 2006). Endodontic irri-

gants must have effective antimicrobial activity but

also exhibit relatively no cytotoxicity toward periapical

and oral mucosal tissue. An anti-inflammatory action

especially in cases of persistent apical periodontitis

might also be advantageous.

Enterococcus faecalis and Candida albicans have been

reported to be of particular interest in cases of persistent

periodontitis (Molander et al. 1998, Sundqvist et al.

1998, Siqueira & Rocas 2004, Fouad et al. 2005).

Anaerobic bacteria, such as Peptostreptococcus micros or

Gram-negative bacteria including P. species have also

been associated with persistent infections (Siqueira

2002). These microorganisms grow in highly resistant

biofilms (Pinheiro et al. 2003), but also as planctonic

cells suspended in the fluid phase of the root canal or as

remnants after mechanical canal preparation (Distel

et al. 2002, Nair 2006).

Sodium hypochlorite (up to 5.25%) is the most

commonly used root canal irrigant and has been used

alternately with H2O2 (3%) (Takeda et al. 1999).

Chlorhexidine digluconate (2%) has also been recom-

mended for root canal irrigation in combination with

mechanical debridement (Siqueira et al. 1998, Gomes

et al. 2001, Basrani & Lemonie 2005). However, the

success rate for conventional treatment of persistent

and refractory apical periodontitis is only in the order of

between 50% and 70% (Weiger et al. 2001) and

consequently NaOCl up to 3% has been reported to

have limited efficacy against high-pathogenic endo-

dontic microorganisms and CHX 2% has demonstrated

inconsistent results (Siqueira et al. 1998, Gomes et al.

2001). In addition, side effects such as haemorrhage,

oedema and skin ulceration have been reported when

high concentrations of NaOCl and H2O2 come into

contact with oral tissues (Pashley et al. 1985, Oncag

et al. 2003, Gernhardt et al. 2004). A significant degree

of cytotoxicity towards oral cells has been found in vitro

as well (Hyslop et al. 1988, Nagayoshi et al. 2004,

Huth et al. 2006). Chlorhexidine (2%) may cause

mucosal desquamation, impaired wound healing and

tooth staining (Bassetti & Kallenberger 1980, Cline &

Layman 1992) and a high cytotoxic potential has been

demonstrated on epithelial cells (Huth et al. 2006).

Therefore, an alternative endodontic antiseptic with

high antimicrobial potential and fewer side effects

would be valuable.

Ozone is currently being discussed as a possible

alternative antiseptic agent in dentistry because of its

reported high antimicrobial power without the devel-

opment of drug resistance (Restaino et al. 1995,

Paraskeva & Graham 2002). Ozone gas in a concen-

tration of �4 g m)3 (HealOzone; KaVo, Biberach,

Germany) is already being used clinically for endo-

dontic treatment. However, results of studies into its

efficacy against endodontic pathogens has been incon-

sistent, and there is little information regarding the

most appropriate application time and concentration to

use (Nagayoshi et al. 2004, Arita et al. 2005, Bezruk-

ova et al. 2005, Hems et al. 2005). Regarding the

demand on relative non-toxicity toward periapical and

oral mucosal tissue for the endodontic irrigants (Nair

2006), the ozone gas concentration currently used in

endodontics (4 g m)3) has been shown to be slightly

less cytotoxic than NaOCl (2.5%) and aqueous ozone

(up to 20 lg mL)1) showed essentially no toxicity to

oral cells in vitro (Filippi 2001, Ebensberger et al. 2002,

Nagayoshi et al. 2004, Huth et al. 2006). The aim of

this study was to investigate the antimicrobial efficacy

of gaseous and aqueous ozone against specific endo-

dontic pathogens in suspension and in biofilms grown

in human root canals.

Materials and methods

Microorganisms

Freeze-dried microorganisms: E. faecalis (ATCC 14506;

LGC Promochem, Wesel, Germany), C. albicans (ATCC

MYA-273), P. micros (ATCC 33270) and P. aeruginosa

(ATCC 15442) were suspended in brain heart infusion

medium (BHI) and recultivated on Schaedler agar

plates (vitamin K1 and 5% sheep blood; BD Diagnostic

Systems, Heidelberg, Germany).

Test agents

Dose–response experiments were performed for gaseous

and aqueous ozone covering a concentration range as

wide as possible to evaluate if there was a concentra-

tion that could possibly compete with the established

endodontic irrigants in antimicrobial effectiveness.

Basically following a log2 scale, the concentration

ranges were limited because of the experimental setting

and equipment. Ozone gas (Ozonosan photonic,

Dr Hansler, Iffezheim, Germany) in concentrations

between 1 g m)3 (the minimum concentration to

measure by the available ozone gas measuring device)

and 53 g m)3 (the highest achievable concentration

because of the experimental set-up and the limitation

Ozone and endodontic biofilms Huth et al.

International Endodontic Journal, 42, 3–13, 2009 ª 2008 International Endodontic Journal4

of the ozone generator) was applied to the test

microorganisms in a self-constructed glass chamber

with simultaneous concentration measurement

(GM-6000-NZL; Anseros, Tubingen, Germany). The

analytical method of the concentration measuring

device is based on UV light absorption at a wavelength

of 253.7 nm where gaseous ozone has its maximum

absorbance (Bocci 2002). For aqueous ozone, bi-

distilled water was treated with ozone gas (75 lg mL)1,

15 min) using the ozone generator, which resulted in a

final photometrically confirmed (Palintest 1000 Ozone

Meter, Palintest Ltd, Gateshead, UK) ozone concentra-

tion in water of 20 lg mL)1 (saturation point), which

was diluted to 1.25 lg mL)1. The ozone concentration

measurement in water involves the oxidation of a

colourless indicator (diethyl-p-phenylene diamine) to

a pink compound by ozone in comparison with a

reference sample without ozone (manufacturer’s infor-

mation, operating wavelength of the photometer is

505 nm). Ozone was compared with freshly prepared

solutions of NaOCl (5.25%, 2.25%), CHX (2%), H2O2

(3%) and phosphate buffered saline (PBS) as a control.

As ozone is an endothermic, highly instable oxygen

compound (Sehested et al. 1991, Hoigne 1998,

Stubinger et al. 2006), both the gas and the ozonated

bi-distilled water were freshly prepared before each

experiment. During production and processing of the

ozone experiments only ozone-resistant materials were

used (e.g. ozone demand-free glass, ozone-resistant

piping material).

Testing ozone against microorganisms in suspension

Microorganisms were grown overnight (37 �C, 10 mL

of BHI), centrifuged, resuspended in PBS to a turbidity

of McFarland 1 [3 · 108 colony forming units

(CFU) mL)1] and diluted 1 : 3. Ten microlitres were

suspended in 1 mL of agent for 1 min followed by

immediate, appropriate dilution with PBS as evaluated

by preceding experiments. Thereof, 10 lL were plated

out on agar plates and incubated aerobically (48 h,

37 �C). For the obligate anaerobic P. micros, all

experimental steps were completed in an anaerobic

work bench (Bactron, Sheldon Manufacturing Inc.,

Cornelius, OR, USA; 85% N2, 5% H2, 10% CO2; 37 �C).Again, 10 lL of an equal dilution of the specific

microorganism suspension were plated out on agar

plates and exposed to ozone gas whilst the control

plates were exposed to ambient air (1 min). After

incubation of the agar plates (48 h, 37 �C), the number

of CFU mL)1 was determined.

Testing ozone against microorganisms in biofilms

grown in human root canals

Crowns of freshly extracted single rooted permanent

teeth (root length 18–19 mm), were removed at the

level of the cemento-enamel junction. The use of the

teeth for these experiments had been agreed upon by

informed consent of the patients. The root canals were

instrumented to the size ISO 40 (K-files; Dentsply

Maillefer, Ballaigues, Switzerland), the apical regions to

size 30 (ProFile�.04; Dentsply Maillefer) with intermit-

tent canal irrigation following each file size (3 mL of

NaOCl 5.25%) (Takeda et al. 1999, Zehnder et al.

2003). Finally, the canals were irrigated with EDTA

10% (5 min, 10–30 mL) followed by normal saline

(Zehnder et al. 2003), dried with paper points and the

roots sterilized (121 �C, 2 bars, 5 min).

The biofilm growth assembly (Fig. 1) contained a

programmable peristaltic pump (IPC-8; Ismatec, Wert-

heim-Mondfeld, Germany), freshly prepared autoclaved

artificial complete saliva (Pratten et al. 1998), 10%

aqueous sucrose solution (Sigma-Aldrich, Schnelldorf,

Germany), flexible silicone tubes (diameter 1 or

2.06 mm; Hartlmaier, Munich, Germany), several

flasks and the prepared dental roots. The ingredients

for the saliva were from Oxoid (Wesel, Germany),

Sigma-Aldrich and BD Diagnostic Systems. All the

equipment was sterilized before use. Overnight cultures

of E. faecalis, C. albicans or P. aeruginosa were used. The

latter species substituted the anaerobic P. micros which

could not be evaluated since the growth assembly was

too large to be incorporated into the anaerobic work

bench. The experiments with P. aeruginosa were con-

fined to the biofilm trials because of the greater

relevance than the suspension experiments already

undertaken for the other three strains.

The artificial saliva was constantly pumped through

a flexible tube into a 50-mL reservoir, supplemented

with the sucrose solution three times a day (30 min,

3 · 33 mL) (Wilson et al. 1998). For the first week, an

overnight culture (37 �C in 10 mL of BHI) of the

respective strain was added daily to the saliva. The

nutrient broth from the reservoir was pumped

(720 mL day)1) (Wilson et al. 1998) through the

canals of four parallel-mounted dental roots each

hanging in a flask, the coronal canal orifice connected

to the flexible tube by a 10-lL micropipette tip

(Eppendorf, Hamburg, Germany). To avoid a contam-

ination of the root surface, the used saliva, which

dropped from the roots’ apical region to the bottom of

the flask, was constantly pumped off into a waste flask

Huth et al. Ozone and endodontic biofilms

ª 2008 International Endodontic Journal International Endodontic Journal, 42, 3–13, 2009 5

via a wider flexible tube (diameter 2.06 mm). After

3 weeks, the roots were removed and cut into 5-mm-

thick horizontal slices, and the apical root portions

were disposed of.

For each test condition, one slice was carefully

transferred to a flask and 1 mL of the test agent added

(four independent trials). For the ozone gas exposure,

two experimental settings were used: setting 1, the

slices were laid flat on glass beads into the gas box, that

the gas streamed over the canal space (resembling

canal areas that are difficult to reach); setting 2, the

slices were positioned upright so as to allow the gas to

stream through the root canal (resembling canal parts

that are easy to reach). After 1 min, the agent was

removed or the slice was removed from the gas box,

1 mL of PBS was immediately added and the slice

vortexed for 1 min (Wilson et al. 1998). Restrained

reactions beyond the 1 min contact time could have

occurred as no chemicals were used to stop the action.

Rather, the vast majority of the test agents were

removed immediately after 1 min and PBS added for

appropriate dilution. Thereafter, 100 lL were plated

out on agar plates and incubated (48 h, 37 �C), andthe CFU per plate were counted. Additionally within

setting 2, ozone gas (4 g m)3) was applied for longer

time intervals, i.e. 2.5, 5 and 10 min. The counted

number of CFU were calculated as a percentage of the

respective control (mean ± SD; n = 3–4).

For each of the independent trials, one slice was

checked for the presence of a biofilm inside the root

canal and for possible microbial contamination of the

outer root surface by scanning electron microscopy

(JSM-35 CF; Jeol, Eching, Germany and SmartSEM;

Zeiss, Oberkochen, Germany).

Statistical methods

As a result of the large number of test agents, the

experiments were conducted in several stages each

with its own control. To compare the antimicrobial

Figure 1 Growth assembly for the mono-species biofilms in root canals. Mono-species biofilms of endodontic pathogens were

grown for 3 weeks in prepared dental root canals of extracted single-rooted permanent teeth. For this purpose, a peristaltic pump

carried pre-warmed artificial saliva supplemented with sucrose and microorganism broth from a reservoir through the roots to a

waste flask at a rate of 720 mL day)1 under aerobic conditions at 37 �C. The right part of the drawing is a magnified view of an

exemplary root hanging in a flask.



activity of the agents, the counted CFU were calculated

in percentage of the respective controls (mean ± SD;

n = 3–4). For all experiments, the absolute numbers of

CFU, the percentage values and the means with

standard deviation of the independent trials are given

in the accompanying Supporting Information. Data

were analysed by one-way anova with Tamhane post

hoc tests to compare independent samples (two-tailed

tests, a-level 0.05) (spss software 12; SPSS Inc.,

Chicago, IL, USA).

Results

Effect of ozone on microorganisms in suspension

Firstly, the effect of aqueous and gaseous ozone on the

specific endodontic pathogens in planctonic culture

was evaluated (see Supporting Information, Tables

A–C). Aqueous ozone completely eliminated E. faecalis

and C. albicans when used in concentrations down to

5 lg mL)1, whereas lower concentrations (2.5 and

1.25 lg mL)1) reduced substantially but did not elim-

inate them totally (Fig. 2a,b). In the case of P. micros,

aqueous ozone down to 2.5 lg mL)1 led to complete

eradication whilst 1.25 lg mL)1 was less effective

(Fig. 2c). In comparison, NaOCl and CHX led to a total

elimination of the tested microorganisms, whereas

H2O2 reduced but did not eliminate them. Ozone gas

in concentrations down to the tested minimum of

1 g m)3 for 1 min almost completely eliminated the

tested strains with a mean reduction of more than 99%

(Fig. 2a–c, Supporting Information, Tables A–C). Sta-

tistically, no differences in effectiveness of the different

agents were seen for E. faecalis (anova, P > 0.05).

Regarding C. albicans, H2O2 and low concentrations of

ozonated water (2.5 and 1.25 lg mL)1) were signifi-

cantly less effective than all other agents (P < 0.05).

Against P. micros, low dose ozonated water

(1.25 lg mL)1) was less effective than the other

antiseptics (P < 0.05).

Establishment of the anatomical biofilm model

The experimental set-up (Fig. 1) allowed the growth of

mono-species biofilms of E. faecalis, C. albicans and P.

aeruginosa over 3 weeks in an anatomically correct form

inside the canal of tooth roots. The roots were sectioned

into horizontal slices before exposure to the gas/agents.

The formation of biofilms was checked for the different

species by SEM of one slice for each independent trial as

well as the outer root surfaces, which showed no

bacterial contamination or biofilm formation (pictures

not shown). P. aeruginosa was substituted for the

anaerobic P. micros because the growth assembly was

too large for the anaerobic chamber.

Effect of ozone on microorganisms in biofilms

The antimicrobial action of ozone against E. faecalis,

C. albicans and P. aeruginosa mono-species biofilms was

tested (see Supporting Information, Tables D–F). Appli-

cation of aqueous ozone for 1 min was dose-depen-

dently effective against the microorganisms, its highest

concentration of 20 lg mL)1 revealing mean CFU

reductions of over 96%, similar to CHX 2% (Fig. 3a–

c, Supporting Information, Tables D–F). Sodium

hypochlorite (5.25%) completely eliminated the micro-

organisms, whilst H2O2 was less effective. In this series

of experiments, ozone gas was applied to the root slices

laying flat in the gas box (setting 1), which revealed a

dose-dependent effectiveness of ozone gas against the

different species (Fig. 3a–c). E. faecalis and C. albicans

was almost eliminated by the highest gas concentration

achievable within the experimental setting (53 g m)3)

(Fig. 3a,b) and P. aeruginosa by the highest and the

second highest concentration (Fig. 3c). Statistically, no

significant differences in effectiveness could be found

between the antiseptics for E. faecalis and C. albicans

(anova, P > 0.05). Against P. aeruginosa, ozone gas

4 g m)3 was significantly less effective than NaOCl,

CHX and ozonated water down to 10 lg mL)1 and

ozonated water 10 lg mL)1 less effective than CHX 2%

(anova, P < 0.05). This was mainly because of a very

small standard deviation in comparison with lower gas

and ozone water concentrations, which showed no

significant differences.

Exposure of the biofilm to ozone gas in a different

setting and with longer contact times

In the following, the experimental conditions were

changed by positioning the slices with E. faecalis

biofilms upright with their cut surfaces in front of the

inlet of the gas box as to allow the gas to stream

through the root canal (setting 2) rather than over the

canal space as in the setting before. Two concentrations

were selected, i.e. one high gas concentration

(32 g m)3) as well as a lower concentration, which is

currently used in dentistry (4 g m)3; HealOzone).

Comparing the outcome of the two settings, the high

gas concentration led to complete eradication of viable

cells after 1 min in the new setting whilst in the old



setting only a reduction was observed (Fig. 4a). Ozone

gas in the lower concentration (1 min) reduced the cell

count more than before, but not to zero. Therefore, as a

last step, the effect of longer exposure times (2.5, 5 and

10 min) of this concentration was tested on the

bacterial biofilms. Contact times of 2.5 min and more

with 4 g m)3 ozone gas led to complete elimination of

the microorganisms (Fig. 4b), but without being sig-

nificantly different to the cell count after 1 min

(P > 0.05) (see Supporting Information, Table G).

Discussion

In this study, gaseous ozone in concentrations down to

1 g m)3 substantially and aqueous ozone down to

5 lg mL)1 completely eliminated the tested planctonic

(a)

(b)

(c)

Figure 2 Antimicrobial efficacy of ozone against endodontic pathogens in suspension. The suspended microorganisms were

exposed to aqueous or gaseous ozone in different concentrations or established endodontic irrigants (NaOCl, CHX and H2O2) for

1 min. The numbers of CFU after contact with PBS for 1 min were defined as 100% control (dotted line). The remaining CFU after

agent/gas exposure were counted and calculated as a percentage of the control (n = 3–4, mean ± SD) (see Supporting

Information, Tables 1–3). (a) shows the antimicrobial activity against E. faecalis. (b) shows the antimicrobial activity against

C. albicans. (c) Shows the antimicrobial activity against P. micros.



pathogens. Gaseous and aqueous ozone were dose- and

strain-dependently effective against the micro-

organisms in biofilms. Total elimination of the

microorganisms in terms of the methods used here

could be achieved by ozone gas at 32 g m)3 for 1 min

or a lower concentration (4 g m)3) for longer contact

times (‡2.5 min) in case of E. faecalis (setting 2).

Aqueous ozone in the highest concentration

(20 lg mL)1, 1 min) nearly eliminated E. faecalis,

C. albicans and P. aeruginosa biofilms.

The root canal model used in these experiments

allowed for the growth of biofilms inside the canal. To

(a)

(b)

(c)

Figure 3 Antimicrobial efficacy of ozone against endodontic pathogens associated in mono-species biofilms in a root canal model.

Mono-species biofilms were grown inside of dental root canals of extracted single-rooted teeth for 3 weeks. In the following, the

roots were cut into horizontal slices that were exposed to aqueous or gaseous ozone in different concentrations or established

endodontic irrigants (NaOCl, CHX, H2O2) or PBS as a control for 1 min. After removal and suspension of the biofilms, the

remaining number of CFU were counted and calculated in % of the CFU counts after contact with the PBS control (100%, dotted

line) (n = 4, mean ± SD) (see Supporting Information, Tables 4–6). (a) Antimicrobial efficacy against E. faecalis biofilm. (b)

Antimicrobial efficacy against C. albicans biofilm. (c) Antimicrobial efficacy against P. aeruginosa biofilm.



determine the efficacy of ozone as alternative antiseptic,

it was compared with traditional endodontic irrigants

(NaOCl, CHX and H2O2) by adding the agents for

1 min. The dose–response experiments for ozone and

additionally the time-course experiments for the ozone

gas concentration currently used in dentistry (4 g m)3,

HealOzone) were aimed at finding a dose-time-concen-

tration that could completely eliminate the microor-

ganisms in the test model as a basis for clinical study

designs in the future. As a source of impreciseness in

the present study, no chemicals were used to arrest the

action of the agents. Therefore, the contact times, e.g.

for CHX, which is known for its substantivity (Khademi

et al. 2006), might be prolonged similar as in the

clinical situation.

Earlier studies reported in part contradictory results

regarding the efficacy of ozone against endodontic

pathogens: one group tested ozonated water

(4 lg mL)1, 10 min) against E. faecalis incubated on

dentine blocks for 6 days (Nagayoshi et al. 2004). A

significant reduction was found but complete elimina-

tion was not observed as was the case with NaOCl

2.5%, which is consistent with the present results.

Additionally, the trials reported here revealed that the

highest concentration of ozonated water (20 lg mL)1)

led to a near eradication of the microorganisms in the

3-week-old biofilm and a complete elimination by

gaseous ozone at a concentration of 32 g m)3 for

1 min or a lower concentration (4 g m)3) for contact

times of at least 2.5 min (setting 2). Further, the biofilm

experiments revealed a near eradication of E. faecalis by

CHX 2% whereas H2O2 was less efficient throughout.

Another study found no significant reduction of

E. faecalis biofilms (grown on membranes for 48 h)

using ozonated water, but did so against planctonic

bacteria (Hems et al. 2005). A reason for these differing

(a)

(b)

Figure 4 Antimicrobial efficacy of ozone gas applied in two experimental settings to E. faecalis biofilms and the effect of prolonged

exposure times. The biofilms were grown as described in Fig. 3. In setting 1, the horizontal root slices were laid flat on glass beads

in the gas box (see experiments in Fig. 3). In setting 2, the horizontal root slices were positioned upright with their cut surfaces in

front of the gas inlet as to allow the gas streaming through the canals. (a) The antimicrobial effect of gaseous ozone in

concentrations of 32 g m)3 and 4 g m)3 for 1 min on the E. faecalis biofilms according to setting 1 (grey bars) and 2 (black bars) is

shown in comparison. PBS served as control. The remaining CFU were counted and calculated in % of the PBS control which was

defined as 100% (dotted line) (n = 3, mean ± SD). (b) The antimicrobial effect of ozone gas (4 g m)3) according to setting 2 for

1 min and prolonged contact times (2.5 min, 5 min and 10 min) is depicted (see Supporting Information, Table 7).



results compared with the study mentioned above

(Nagayoshi et al. 2004) and the present experiments

revealing a CFU reduction when exposed to high

concentrated ozonated water might be that a rather

low ozone concentration was used in the other study

(Hems et al. 2005). That is, ozone gas was bubbled

through the water containing the biofilm for only

4 min. The maximum concentration of ozonated water

(20 lg mL)1) was achieved in the present study only

after 15 min of ozonation (data not shown). Another

recent study grew E. faecalis biofilms over 60 days in

root canals and applied ozonated water, ozone gas,

NaOCl 2.5% or CHX 2% for 20 min (Estrela et al.

2007). Contrasting to the present results, none of the

irrigants were found to have an antimicrobial effect.

The effect of ozone against C. albicans has been

reported primarily for denture cleaning (Murakami

et al. 1996, Oizumi et al. 1998). More recently,

C. albicans incubated on resin plates for 120 min was

almost eliminated by use of ozonated water (2 and

4 lg mL)1, 1 min) with or without ultrasonication

(Arita et al. 2005). As 120 min represents a short time

interval for biofilm formation, that study might be

better compared with the present suspension experi-

ments, in which a mean reduction of about 86% of

C. albicans by 2.5 lg mL)1 ozonated water and a total

elimination by 5 lg mL)1 ozonated water and a

reduction of over 99% by ozone gas down to 1 g m)3

was achieved. In the present biofilm experiments,

C. albicans was found to be completely eliminated only

by NaOCl (5.25%) and to over 96% by 53 g m)3

gaseous ozone (setting 1), 20 lg mL)1 ozonated water

and CHX 2%.

The effect of ozone against the anaerobe P. micros

has not been evaluated before. Ozone gas in the tested

minimum concentration (1 g m)3, setting 1) and

aqueous ozone (‡ 2.5 lg mL)1) completely eliminated

the suspended microorganisms. Biofilm experiments

were not performed with P. micros as the growth

assembly could not be maintained in anaerobic condi-

tions.

The use of ozone as a disinfectant against

P. aeruginosa in dental unit water lines has been

reported, but there is no information about the required

time and concentration for total elimination (Filippi

1995, Al Shorman et al. 2003). In present biofilm

experiments, total eradication was achieved by ozone

gas concentrations of 32 g m)3 (setting 1) and NaOCl

(2.25%, 1 min). High-concentrated aqueous ozone

(20 lg mL)1, 1 min) and CHX 2% almost eliminated

the viable microorganisms.

Conclusions

High-concentrated gaseous and aqueous ozone was

dose-, strain- and time-dependently effective against

the tested microorganisms in suspension and the biofilm

test model. However, NaOCl was the only method that

completely eliminated all types of microorganisms.

Acknowledgements

The authors wish to acknowledge E. Thielke and

C. Kohler for technical project support. The study was

financed by the Medical Faculty, University of Munich

(FoFoLe Reg. Nr. 401), departmental funding and the

KaVo Company.

Supporting information

Additional supporting information may be found in the

online version of this article:

Table S1 Antimicrobial efficacy of ozone and estab-

lished endodontic irrigants (1 min) against the tested

microorganisms in suspension or associated in biofilms.

The absolute number of remaining colony forming units

(CFU abs) of 3 to 4 independent trials (n = 3–4) are

given. The CFU are also given in % of the respective

controls in parentheses and their means with the

standard deviations of the independent trials (% control,

mean ± SD) which correspond to Fig. 2, 3 and 4. NaOCl,

sodium hypochlorite; CHX, chlorhexidine digluconate;

H2O2, hydrogen peroxide; O3, ozone. The antimicrobial

effects against E. faecalis (A), C. albicans (B), and P. micros

(C) in suspension are shown as well as against E. faecalis

(D), C. albicans (E) and P. aeruginosa (F) associated in

biofilms. Table G shows the antimicrobial efficacy of

ozone gas in concentrations of 32 g/m3 and 4 g/m3

applied in two experimental settings to E. faecalis biofilms

and the effect of prolonged exposure times in setting 2

(1 min, 2.5 min, 5 min, 10 min).

Please note: Wiley-Blackwell are not responsible for the

content or functionality of any supporting information

supplied by the authors. Any queries (other thanmissing

material) should be directed to the corresponding author

for the article.

References

Al Shorman H, Abu-Naba¢a L, Coulter WA, Lynch E (2003)

Primary colonization of DUWL by P. aeruginosa and its

eradication by ozone (Abstract). Journal of Dental Research

82, B 284.



Arita M, Nagayoshi M, Fukuizumi T et al. (2005) Microbicidal

efficacy of ozonated water against Candida albicans adhering

to acrylic denture plates. Oral Microbiology and Immunology

20, 206–10.

Basrani B, Lemonie C (2005) Chlorhexidine Gluconate.

Australian Endodontic Journal 31, 48–52.

Bassetti C, Kallenberger A (1980) Influence of chlorhexidine

rinsing on the healing of oral mucosa and osseous lesions.

Journal of Clinical Periodontology 7, 443–56.

Bezrukova IV, Petrukhina NB, Voinov PA (2005) Experience

in medical ozone use for root canal treatment. Stomatologiia

84, 20–2.

Bocci V (2002) Oxygen-Ozone Therapy. A Critical Evaluation, 1st

edn. Dordrecht, The Netherlands: Kluwer Academic Pub-

lishers.

Cline NV, Layman DL (1992) The effects of chlorhexidine on

the attachment and growth of cultured human periodontal

cells. Journal of Periodontology 63, 598–602.

Distel JW, Hatton JF, Gillespie MJ (2002) Biofilm formation in

medicated root canals. Journal of Endodontics 28, 689–93.

Ebensberger U, Pohl Y, Filippi A (2002) PCNA-expression of

cementoblasts and fibroblasts on the root surface after

extraoral rinsing for decontamination. Dental Traumatology

18, 262–6.

Estrela C, Estrela CRA, Decurcio DA, Hollanda ACB, Silva JA

(2007) Antimicrobial efficacy of ozonated water, gaseous

ozone, sodium hypochlorite and chlorhexidine in infected

human root canals. International Endodontic Journal 40, 85–

93.

Filippi A (1995) Efficacy of disinfecting water in dental

treatment units using ozone. Deutsche Zahnarztliche Zeitsch-

rift 50, 708–10 (in German).

Filippi A (2001) The effects of ozonized water on epithelial

wound healing (in German). Deutsche Zahnarztliche Zeitsch-

rift 56, 104–8.

Fouad AF, Zerella J, Barry J, Spangberg LS (2005) Molecular

detection of Enterococcus species in root canals of therapy-

resistant endodontic infections. Oral Surgery, Oral Medicine,

Oral Pathology, Oral Radiology and Endodontics 99, 112–8.

Gernhardt CR, Eppendorf K, Kozlowski A, Brandt M (2004)

Toxicity of concentrated sodium hypochlorite used as an

endodontic irrigant. International Endodontic Journal 37,

272–80.

Gomes BPFA, Ferraz CCR, Vianna ME, Berber VB, Teixeira FB,

Souza-Filho FJ (2001) In vitro antimicrobial activity of

several concentrations of sodium hypochlorite and chlorh-

exidine gluconate in the elimination of Enterococcus faecalis.

International Endodontic Journal 34, 424–8.

Hems RS, Gulabivala K, Ng Y-L, Ready D, Spratt DA (2005)

An in vitro evaluation of the ability of ozone to kill a strain of

Enterococcus faecalis. International Endodontic Journal 38, 22–

9.

Hoigne J (1998) Chemistry of aqueous ozone, and transfor-

mation of pollutants by ozonation and advanced oxidation

processes. In: Hubrec J, ed. The Handbook of Environmental

Chemistry Quality and Treatment of Drinking Water. Berlin,

Germany: Springer, pp. 83–141.

Huth KC, Jakob FM, Saugel B et al. (2006) Effect of ozone on

oral cells compared with established antimicrobials. Euro-

pean Journal of Oral Sciences 114, 435–40.

Hyslop PA, Hinshaw DB, Halsey WA et al. (1988) Mechanisms

of oxidant-mediated cell injury. The glycolytic and mito-

chondrial pathways of ADP phosphorylation are major

intracellular targets inactivated by hydrogen peroxide.

Journal of Biological Chemistry 263, 1665–75.

Khademi AA, Mohammadi Z, Havaee A (2006) Evaluation of

the antibacterial substantivity of several intra-canal agents.


Molander A, Reit C, Dahlen G, Kvist T (1998) Microbiological

status of root-filled teeth with apical periodontitis. Interna-

tional Endodontic Journal 31, 1–7.

Murakami H, Sakuma S, Nakamura K et al. (1996) Disinfec-

tion of removable dentures using ozone. Dental Materials

Journal 15, 220–5.

Nagayoshi M, Kitamura C, Fukuizumi T, Nishihara T,

Terashita M (2004) Antimicrobial effect of ozonated water

on bacteria invading dentinal tubules. Journal of Endodontics

30, 778–81.

Nair PN (2006) On the causes of persistent apical peri-

odontitis: a review. International Endodontic Journal 39,

249–81.

Oizumi M, Suzuki T, Uchida M, Furuya J, Okamoto Y (1998)

In vitro testing of a denture cleaning method using ozone.

Journal of Medical and Dental Sciences 45, 135–9.

Oncag O, Hosgor M, Hilmioglu S, Zekioglu O, Eronat C,

Burhanoglu D (2003) Comparison of antimicrobial and

toxic effects of various root canal irrigants. International

Endodontic Journal 36, 423–32.

Paraskeva P, Graham NJD (2002) Ozonation of municipal

wastewater effluents. Water Environment Research 74, 569–

81.

Pashley EL, Birdsong NL, Bowman K, Pashley DH (1985)

Cytotoxic effects of NaOCl on vital tissue. Journal of

Endodontics 11, 525–8.

Pinheiro ET, Gomes BPFA, Ferraz CCR, Sousa ELR, Teixeira

FB, Souza-Filho FJ (2003) Microorganisms from root canals

of root-filled teeth with periapical lesions. International


Pratten J, Wills K, Barnett P, Wilson M (1998) In vitro studies

of the effect of antiseptic-containing mouthwashes on the

formation and viability of Streptococcus sanguis biofilms.

Journal of Applied Microbiology 84, 1149–55.

Restaino L, Frampton EW, Hemphill JB, Palnikar P (1995)

Efficacy of ozonated water against various food-related

microorganisms. Applied and Environmental Microbiology 61,

3471–5.

Sehested K, Corfitzen H, Holcman J, Fischer CH, Hart EJ (1991)

The primary reaction in the decomposition of ozone in acidic

aqueous solutions. Environmental Science & Technology 25,

1589–96.



Siqueira JF Jr (2002) Endodontic infections: concepts,

paradigms, and perspectives. Oral Surgery, Oral Medicine,

Oral Pathology, Oral Radiology and Endodontics 94,

281–93.

Siqueira JF Jr, Rocas IN (2004) Polymerase chain reaction-

based analysis of microorganisms associated with failed

endodontic treatment. Oral Surgery, Oral Medicine, Oral

Pathology, Oral Radiology and Endodontics 97, 85–94.

Siqueira JF Jr, Batista MMD, Fraga RC, de Uzeda M (1998)

Antimicrobial effects of endodontic irrigants on black-

pigmented gram-negative anaerobes and facultative bacte-

ria. Journal of Endodontics 24, 414–6.

Stubinger S, Sader R, Filippi A (2006) The use of ozone in

dentistry and maxillofacial surgery: a review. Quintessence

International 37, 353–9.

Sundqvist G, Figdor D, Persson S, Sjogren U (1998) Microb-

iologic analysis of teeth with failed endodontic treatment

and the outcome of conservative re-treatment. Oral Surgery,

Oral Medicine, Oral Pathology, Oral Radiology and Endodontics

85, 86–93.

Takeda FH, Harashima Y, Kimura Y, Matsumoto K (1999) A

comparative study of the removal of smear layer by three

endodontic irrigants and two types of laser. International


Weiger R, Geurtsen W, Heidemann D et al. (2001) About the

prognosis of root canal treatments. National Statement of

the DGZMK (in German). Deutsche Zahnarztliche Zeitschrift

56, 206–7.

Wilson M, Patel H, Noar JH (1998) Effect of chlorhexidine on

multi-species biofilms. Current Microbiology 36, 13–8.

Zehnder M, Lehnert B, Schonenberger K, Waltimo T (2003)

Irrigating solutions and intracanal medicaments in end-

odontics (in German). Schweizerische Monatsschrift fur Zahn-

medizin 113, 756–63.



Stress distribution of three NiTi rotary files underbending and torsional conditions using amathematic analysis

T. O. Kim1, G. S. P. Cheung2, J. M. Lee3, B. M. Kim3, B. Hur1 & H. C. Kim1

1Department of Conservative Dentistry, School of Dentistry, Pusan National University, Busan, Korea; 2Area of Endodontics,

Faculty of Dentistry, The University of Hong Kong, Hong Kong; and 3Division of Precision Manufacturing Systems, Pusan National

University, Busan, Korea

Abstract

Kim TO, Cheung GSP, Lee JM, Kim BM, Hur B, Kim HC.

Stress distribution of three NiTi rotary files under bending and

torsional conditions using a mathematic analysis. International

Endodontic Journal, 42, 14–21, 2009.

Aim To compare and evaluate the stress distribution

of three NiTi instruments of various cross-sectional

configurations under bending or torsional condition

using a finite-element analysis model.

Methodology Three NiTi files (ProFile, ProTaper

and ProTaper Universal) were scanned using Micro-CT

to produce a three-dimensional digital model. The

behaviour of the instrument under bending or torsional

loads was analysed mathematically in software (ABA-

QUS V6.5-1), taking into consideration the nonlinear

mechanical characteristic of NiTi material.

Results ProFile showed the greatest flexibility, fol-

lowed by ProTaper Universal and ProTaper. The

highest stress was observed at the surface near the

cutting edge and the base of (opposing) flutes during

cantilever bending. Concentration of stresses was

observed at the bottom of the flutes in ProFile and

ProTaper Universal instruments in torsion. The stress

was more evenly distributed over the surface of

ProTaper initially, which then concentrated at the

middle of the convex sides when the amount of angular

deflection was increased.

Conclusion Incorporating a U-shaped groove in the

middle of each side of the convex-triangular design

lowers the flexural rigidity of the origin ProTaper

design. Bending leads to the highest surface stress at or

near the cutting edge of the instrument. Stress

concentration occurs at the bottom of the flute when

the instrument is subjected to torsion.

Keywords: bending, cross-sectional geometry, finite-

element analysis, NiTi rotary file, stress distribution,

torsion.

Received 19 March 2008; accepted 16 August 2008

Introduction

Root canal instruments manufactured with nickel–

titanium (NiTi) have been developed in an attempt to

overcome the rigidity of instruments made from stain-

less steel alloys (Walia et al. 1988). NiTi instruments

possess a lower modulus of elasticity and a superior

resistance to torsional fracture, compared with stainless

steel instruments of similar size (Walia et al. 1988,

Schafer et al. 2003). The NiTi rotary instruments allow

root canal preparation to be accomplished more expe-

ditiously than hand instruments; a well-centred,

tapered root canal form with minimal risk of trans-

porting the original canal centre is often achieved

(Glosson et al. 1995, Garip & Gunday 2001, Schafer

2001, Chen & Messer 2002, Lee et al. 2003, Schafer

et al. 2004).

To date, many NiTi rotary systems have been

introduced to the market. Most brands, e.g. ProFile

(Dentsply Maillefer, Ballaigues, Switzerland), K3

(SybronEndo, Orange, CA, USA), Mtwo (VDW, Munich,

Germany) and Hero Shaper (Micro-Mega, Besancon,

Correspondence: Dr Hyeon-Cheol Kim, DDS, MS, PhD, Assis-

tant Professor, Department of Conservative Dentistry, Pusan

National University School of Dentistry, 1-10, Ami-dong,

Seo-gu, Busan 602-739, Korea (Tel.: +82 51 240 7978;

fax: +82 51 254 0575; e-mail: [email protected]).

doi:10.1111/j.1365-2591.2008.01481.x


France) have a regularly tapered shaft, but with

different cross-sectional designs; some also possess

‘radial lands’ (Schafer 2001, Hata et al. 2002).

Amongst these systems, the ProFile system is best

known for its U-file design (i.e. with a concave, ‘U-

shaped’ flutes in cross-section; Fig. 1a), and for its

flexibility and better centering ratio than some other

systems (Park et al. 2003, Walsch 2004, Kim et al.

2005). In contrast, the ProTaper system (Dentsply

Maillefer) has a unique design for its shaft with a

‘progressively changing’ taper (Bergmans et al. 2003,

Clauder & Baumann 2004). The original cross-sec-

tional configuration of the ProTaper system was

triangular with convex sides (Fig. 1b). The sharp

cutting edge (instead of a radial land) is claimed to

reduce the contact area between the file and dentine,

thus enhancing the cutting efficiency of the instrument

(Clauder & Baumann 2004). However, it has been

reported that the ProTaper system tends to produce

more aberrations, transportation or straightening of

the canal (Yun & Kim 2003, Calberson et al. 2004,

Schafer et al. 2004). To overcome the problem, which

Figure 1 Schematic drawings of the cross-sectional and longitudinal geometry of three NiTi files after the real-size, three-

dimensional image from micro-CT: (a) ProFile size 30, 0.06 taper; (b) ProTaper F3 and (c) ProTaper Universal F3.

Kim et al. Stress distribution of NiTi rotary files


may be related to a slightly greater rigidity (partly

because of the cross-sectional area; another factor

being the taper of the instrument), compared with

ProFile instruments of similar cross-sectional dimen-

sion, a new version with a modified cross-sectional

design for the larger instruments of the original system

has been marketed as ProTaper Universal. The F2 and

F3 instruments of the ProTaper Universal system have

incorporated an additional groove in the middle of each

side of the ‘convex-triangular’ cross-section in an

attempt to increase its flexibility (Fig. 1c).

Clinically, there is a potential risk of rotary NiTi

instruments fracturing in the canal – even new

instrument may demonstrate unexpected failure in

use (Arens et al. 2003). On the other hand, little is

known about the distribution of stresses, an important

factor related to instrument fracture, when the instru-

ment is subjected to bending or torsional load. It has

been reported that fracture of an engine-file may occur

in either one or a combination of two ways: torsional

and flexural (i.e. fatigue) (Sattapan et al. 2000, Cheung

et al. 2005, Wei et al. 2007); the geometrical design is

an important determinant because of the effect on the

torsional and bending properties of the instrument

(Camps et al. 1995). Several studies of the stresses

generated in NiTi instrument have been completed

using finite-element (FE) analysis (Turpin et al. 2000,

2001, Berutti et al. 2003); however, they evaluated a

simulated, cylindrical shape and ignored the taper of

the root canal instrument when constructing the

models. Recently, Xu et al. (2006) have reported on

the effect of cross-section configuration on the mechan-

ical behaviour of root canal files by examining an

idealized cross-sectional configuration with FE analysis,

but they did not seem to have verified the actual

geometry of the real product. Indeed, there could be

discrepancies between the idealized design and the

actual product (Low et al. 2006). Thus, the purpose of

this study was to compare the stress distribution of the

two ProTaper designs under bending and torsional

stresses by inputting the actual shape of the instru-

ments for three-dimensional (3D) FE analysis. A U-file

design (ProFile) was also examined as a control.


Modeling of NiTi rotary file

Real-size, digitized models of three brands of NiTi

instrument: ProFile size 30 (0.06 taper), ProTaper F3

and ProTaper Universal F3 (all from Dentsply Maillefer)

were obtained by first scanning them at 2-lm intervals

in a micro-CT scanner (HMX; X-Tek Group, Santa

Clara, CA, USA). Then, the outline of the instrument

was extracted from the stacks of 3D data in software

(IDEAS11 NX; UGS, Plano, TX, USA). Finally, a mesh of

linear, eight-noded, hexahedral elements was overlaid

onto the rendered 3D image. Such a 3D model

consisted of 11880 elements with 16318 nodes for

ProFile, 7560 elements with 9017 nodes for ProTaper,

or 8964 elements with 10668 nodes for ProTaper

Universal (Fig. 1). This numerical model of each

instrument was entered into a 3D FE analysis package

(ABAQUS V6.5-1; SIMULIA, Providence, RI, USA) with

the z-axis running from the tip to the shaft of the

instrument.

A nonlinear, stress–strain behaviour of the NiTi

material (Wang 2007) was entered for the NiTi

material during the mathematical analysis (Fig. 2):

OA represents the elastic deformation of austenite,

AB the pseudoelastic range (plateau spanning over

about 4% strain) because of stress-induced martensitic

(SIM) transformation, BC the elastic deformation of

martensite, and CD the plastic deformation of the

transformed martensite. Plastic deformation (a result of

because of crystallographic slip) is unrecoverable,

whereas elastic and SIM transformation strains are

mostly recoverable (Duerig & Pelton 1994). The

Young’s modulus of the alloy was 36 GPa and the

Figure 2 Stress–strain relationship of the NiTi material (from

Wang 2007).

Stress distribution of NiTi rotary files Kim et al.


Poisson’s ratio 0.3. The critical stress at the beginning

of the SIM phase transformation was chosen to be

504 MPa and that at the end was 755 MPa (Wang

2007).

Experimental conditions of simulation

The behaviours of the three instruments were analysed

numerically under the following simulated conditions

in the FE analysis (Fig. 3):

1. Cantilever bending with a constant load – deforma-

tion in the form of cantilever bending was simulated by

applying a concentrated load of 1 N at the tip of the file

with its shaft rigidly held in place (Fig. 3a). The vertical

displacement was measured and the von Mises stress

distribution was evaluated.

2. Stress distribution under cantilever bending at fixed

displacement – under a similar condition as (a) above,

the tip of the file was deflected for a distance of 2 mm

(Fig. 3b) and held there. The von Mises stress distribu-

tion was examined.

3. Application of a shear moment (torsion) – a

2.5 Nmm moment of force was applied to the shaft in

a clockwise direction normal to the long axis of the

instrument (Fig. 3c), whilst 4 mm of the tip was rigidly

constrained. The stress distribution was evaluated.

4. Stress distribution at a fixed angular deflection – the

von Mises stress distribution over the instrument was

examined after the instrument was rotated by 10�clockwise with its tip rigidly fixed at 4 mm (Fig. 3d).

Results

Cantilever bending

At a concentrated load of 1 N, the end deflection for

ProFile was 4.6 mm, ProTaper 2.5 mm and ProTaper

Universal 3.1 mm, indicating a greater flexibility for

ProFile instrument. A maximum von Mises stress of

577 MPa was found at 8.4 mm from the tip of the

ProFile instrument; the values were 349 MPa at

3.7 mm for ProTaper, and 547 MPa at 3.6 mm for

ProTaper Universal (Fig. 4). The bending force required

to deflect the instrument from its resting position was

greatest for ProTaper, followed by ProTaper Universal

and ProFile (Fig. 5a). For the same amount of end

deflection (2 mm), a maximum von Mises stress of

387 MPa was noted for ProTaper Universal, again, at

3.6 mm from the instrument tip. The values were

350 MPa at 3.7 mm for ProTaper, and 275 MPa at

8.4 mm for ProFile instrument respectively (Fig. 6a).

The highest stress was observed at the surface at the

cutting edge of ProTaper, but at a very short distance

from such edge for ProTaper Universal and ProFile, and

at the base of the opposing flute during cantilever

bending.

Figure 3 Simulated conditions applied in this study: (a)

cantilever bending with a concentrated load of 1 N applied

to the tip of the instrument; (b) cantilever bending until the tip

was displaced by 2 mm; (c) Shear moment of 2.5 Nmm

applied to the shaft, with the instrument rigidly fixed at 4 mm

from its tip and (d) Similar condition as (c) but with the torque

applied until the shaft was rotated by 10�.

(a)

(b)

(c)

Figure 4 Relative deflection (to scale) of the tip, and stress

distribution under cantilever loading (1 N applied to the tip)

for each instrument: (a) ProFile; (b) ProTaper and (c) ProTaper

Universal.



Shear moment (torsion)

When a torque of 2.5 Nmm was applied, the original

ProTaper design showed the lowest value (350 MPa)

for the maximum von Mises stress, followed by

384 MPa for ProTaper Universal (Fig. 6b). The ProFile

showed the highest stress of 455 MPa, running along

at the base (bottom) of the U-shaped flutes. The angular

deflection was 0.691, 0.826 and 0.995 degrees for

ProTaper, ProTaper Universal and ProFile respectively.

The resistance to torsion mirrored the flexural

rigidity of the instrument: a higher torque was required

to angularly deflect the ProTaper than the other two

instruments (Fig. 5b). The highest von Mises stress

(constrained region not compared) recorded for ProFile

was 333 MPa, ProTaper 359 MPa and ProTaper

Universal 388 MPa, all situated at the base of the

flutes in cross-section (Fig. 6c).

Discussion

In the last decade, the use of NiTi rotary instruments

has grown in popularity and there has been an

increasing number of proprietary systems introduced

commercially. NiTi engine-files operate by way of

continuous rotation in the root canal and, as such,

are subjected to unidirectional torque (assuming no

stalling). The value of the shear (torsional) stress varies

depending on the canal size (Hubscher et al. 2003,

Peters et al. 2003), hardness of the dentine to be cut

(Berutti et al. 2003) and the use of a lubricant (Boessler

Figure 5 (a) Bending moment needed to deflect the tip and (b)

the torque required to rotate each file under the restrained

condition.

(a)

(b)

(c)

Figure 6 Distribution of von Mises stresses

under various conditions for the three

instruments tested, the maximum stress

values (in MPa) for each case being: (a)

ProFile 275, ProTaper 350, ProTaper

Universal 387; (b). ProFile 455, ProTaper

350, ProTaper Universal 384 and (c) ProFile

333, ProTaper 359, ProTaper Universal 388.



et al. 2007). The cross-sectional configuration is also

an important determinant of the distribution of stresses

on the instrument (Tripi et al. 2006). To avoid dimen-

sional discrepancy, the three brands of NiTi instrument

examined in this present study were first scanned to

obtain a real representation of the 3D shape prior to

entry into the mathematical simulation.

Studies of NiTi instrument breakage are usually

completed by means of post-mortem SEM examination

of the fracture mode after clinical or simulated use.

Such evaluation would not reveal the stresses on the

instrument during bending or rotation. Based on a

mathematical comparison of the behaviour of two

theoretical cross-sections of ProTaper and ProFile, it

has been reported that ProTaper might be more

suitable for enlarging the (coronal portion of) canals

during the initial phase of shaping, and that ProFile

might be more suitable for wider canals and in the final

phase of shaping (Berutti et al. 2003). Turpin et al.

(2000) have studied the influence of the idealized cross-

sectional profile (ProFile vs. Hero) on the torsional and

bending stresses using a boundary integral method,

and also suggested that instruments of different cross-

sectional design should be used for different procedures.

The amount of end deflection under cantilever

loading is a measure of the instrument’s flexural

rigidity, the product of the elastic modulus of the

material and second moment of inertia of the part

(Timoshenko & Goodier 1970). ProFile had a greater

deflection than other systems, indicating that ProFile

possesses a lower flexural rigidity, i.e. higher flexibility.

As the mechanical property of the raw material is the

same for the three designs (from the same manufac-

turer), the difference in flexural rigidity of the various

makes is a result of the different geometry. ProTaper

had the greatest flexural rigidity, lower end deflection,

and the least concentration of stress over the surface

when subjected to a load of 1 N. Berutti et al. (2003)

have also reported that ProTaper had lower and more

evenly distributed stresses, compared with the ProFile

model, under similar type of loading. However, in the

clinical situation, the stress generated in an instrument

arises from it having to conform to the root canal

curvature (i.e. fixed deflection) but not due to an

externally applied force. Thus, the situation where the

various brands were subjected to the same amount of

end deflection (i.e. Fig. 3b) would be more relevant

than application of an arbitrary load – both the

ProTaper and ProTaper Universal showed a greater

value of internal stresses than ProFile. The highest

stress concentration was found at the cutting edge of

ProTaper and ProTaper Universal, and near the cutting

edge of the ProFile, and at the bottom of the directly

opposite flute (see Fig. 6a). This is expected from the

mechanics of bending a beam of triangular cross-

section. Generally, flexural (bending) deflection is

proportional to the bending moment and inversely

proportional to sectional modulus (Timoshenko &

Goodier 1970). A correlation between stiffness of an

instrument and its cross-sectional area has been

suggested in many studies (Haıkel et al. 1999, Turpin

et al. 2000, Schafer et al. 2003). In view of the similar

longitudinal outline of the ProTaper and ProTaper

Universal instrument, the addition of a groove (flute) at

the centre of each side of the ‘convex-triangular’ cross-

section has effectively reduced the second moment of

inertia for the latter. On the other hand, this groove

seems to have served as a stress-raiser in torsion.

The torsional rigidity, which is proportional to the

applied torque and the polar moment of inertia of the

part, was evaluated in the present study by measuring

the angular deflection of the instrument. ProTaper was

the most rigid, whereas ProFile the least. However,

unlike bending of the instrument being governed by the

canal curvature, shear stresses are generated in an

engine-file because of friction and the (resistance of

dentine to) cutting action. Thus, it would be more

logical to examine the stress distribution under a similar

torsional moment (Fig. 6b) rather than at the same

twist-angle (Fig. 6c). It seems that ProFile is going to

experience a much greater stress than ProTaper

instrument in such a situation (see Fig. 6c), a finding

corrobating that of other studies using FE analysis

(Turpin et al. 2000, 2001, Berutti et al. 2003, Xu et al.

2006). Concentrations of (torsional) stress were ob-

served at the bottom of the U-shaped flutes for ProFile

and at the concave groove at each side of the triangular

cross-section for ProTaper Universal, the stress of

which was much higher than that for the original

ProTaper. Hence, there is a greater chance of SIM

transformation, and even plastic deformation of the

transformed martensite there. This may explain a

higher reported incidence of unwinding defects (with

or without breakage) for discarded, clinically used,

engine-driven ProFile than ProTaper (Shen et al. 2006)

or K3 instrument (Ankrum et al. 2004). Enlarging the

canal to a size of 15 or 20 before using the instrument

would help to reduce the torsional stress experienced by

the instrument (Hubscher et al. 2003) and lower the

risk of shear fracture. Incorporating a U-shaped groove

for the original ProTaper design, i.e. ProTaper Uni-

versal, would lead to some stress concentration at the



bottom of the groove, as expected. It would be a weaker

point than with the ProTaper, but still be better than

ProFile in strength in order to resist torsion.

The reaction stresses in an instrument (of the same

material and dimensions) are dependent on the

geometry of the working part relative to the operating

load. Factors affecting the stress distribution include

the cross-sectional configuration, the depth of the

flute, area of the inner core and (the bulk of)

peripheral mass in cross-section; all these influence

the magnitude of the second and polar moments of

inertia. Not one of the systems studied was both

highly flexible and yet able to withstand and distribute

the stress evenly in bending and torsion. Indeed, it is

obvious that different parameters are operating when

the fracture susceptibility of an instrument (because of

torsion vs. rotational bending) is concerned (Cheung

et al. 2005). Clinicians should understand not only

the general guidelines for NiTi rotary instrumentation,

but also the structural characteristics which might

influence the durability or the risk of an engine-file to

fracture. To increase safety, endodontic educators

must emphasize the need for mastering the skill for

rotary instruments through appropriate, supervised

training (Mandel et al. 1999, Yared et al. 2001).

Despite a truer representation of the actual geometry

of the instrument in this study, the actual stresses

may differ when the instrument is actively filing

against the dentine wall during clinical use. Further

studies through other methods to verify the relation-

ship between instrument design, stress distribution,

fatigue fracture and the influence of microscopic

notches, are required.

Conclusions

This study examined the stress distribution under

bending or torsional load using a 3D FE analysis for

three NiTi instruments of various cross-sectional con-

figurations. It is concluded that the U-file design had

the lowest flexural rigidity, compared with a ‘convex-

triangular’ cross-section with or without an additional

flute, but a higher magnitude of stress concentration at

the bottom of the flute in torsion. Bending led to the

highest surface stress at or near the cutting edge of all

three instruments. The convex-triangular cross-section

was able to distribute the shear stresses initially, but

had similar stress concentrations at the same degree of

angular deflection. Incorporating a U-shaped groove for

the ProTaper design results in an instrument with

intermediate properties between the two.

References

AnkrumMT, Hartwell GR, Trutt JE (2004) K3 Endo, ProTaper,

and ProFile systems: breakage and distortion in severely

curved root of molars. Journal of Endodontics 30, 234–7.

Arens FC, Hoen MM, Steiman HR, Dietz GC Jr (2003)

Evaluation of single-use rotary nickel–titanium instruments.

Journal of Endodontics 29, 664–6.

Bergmans L, Van Cleynenbreugel J, Beullens M, Wevers M,

Van Meerbeek B, Lambrechts P (2003) Progressive versus

constant tapered shaft design using NiTi rotary instruments.


Berutti E, Chiandussi G, Gaviglio I, Ibba A (2003) Comparative

analysis of torsional and bending stresses in two mathe-

matical models of nickel–titanium rotary instruments:

ProTaper versus ProFile. Journal of Endodontics 29, 15–9.

Boessler C, Peters OA, Zehnder M (2007) Impact of lubricant

parameters on rotary instrument torque and force. Journal of


Calberson FL, Deroose CA, Hommez GM, De Moor RJ (2004)

Shaping ability of ProTaper nickel–titanium files in simu-

lated resin root canals. International Endodontic Journal 37,

613–23.

Camps JJ, Pertot WJ, Levallois B (1995) Relationship between

file size and stiffness of nickel titanium instruments.

Endodontics and Dental Traumatology 11, 270–3.

Chen JL, Messer HH (2002) A comparison of stainless steel

hand and rotary nickel–titanium instrumentation using a

silicone impression technique. Australian Dental Journal 47,

12–20.

Cheung GS, Peng B, Bian Z, Shen Y, Darvell BW (2005) Defects

in ProTaper S1 instruments after clinical use: fractographic

examination. International Endodontic Journal 38, 802–9.

Clauder T, Baumann MA (2004) ProTaper NT system. Dental

Clinics of North America 48, 87–111.

Duerig TW, Pelton AR (1994) Ti–Ni shape memory alloys. In:

Boyer R, Welsch G, Collings EW, eds. Materials Properties

Handbook: Titanium Alloys. Materials Park: ASM Interna-

tional, pp. 1035–48.

Garip Y, Gunday M (2001) The use of computed tomography

when comparing nickel–titanium and stainless steel files

during preparation of simulated curved canals. International


Glosson CR, Haller RH, Dove SB, del Rio CE (1995) A

comparison of root canal preparations using Ni–Ti hand,

Ni–Ti engine-driven, and K-Flex endodontic instruments.


Haıkel Y, Serfaty R, Bateman G, Senger B, Allemann C (1999)

Dynamic and cyclic fatigue of engine-driven rotary nickel–

titanium endodontic instruments. Journal of Endodontics 25,

434–40.

Hata G, Uemura M, Kato AS, Imura N, Novo NF, Toda T

(2002) A comparison of shaping ability using ProFile, GT

file, and Flex-R endodontic instruments in simulated canals.




Hubscher W, Barbakow F, Peters OA (2003) Root canal

preparation with FlexMaster: assessment of torque and force

in relation to canal anatomy. International Endodontic Journal

36, 883–90.

Kim HC, Park JK, Hur B (2005) Relative efficacy of three Ni-Ti

file systems used by undergraduates. Journal of Korean

Academy of Conservative Dentistry 30, 38–48.

Lee CH, Cho KM, Hong CU (2003) Effect of various canal

preparation techniques using rotary nickel–titanium files on

the maintenance of canal curvature. Journal of Korean

Academy of Conservative Dentistry 28, 41–9.

Low D, Ho AW, Cheung GS, Darvell BW (2006) Mathematical

modeling of flexural behavior of rotary nickel titanium

endodontic instruments. Journal of Endodontics 32, 545–8.

Mandel E, Adib-Yazdi M, Benhamou LM, Lachkar T, Mesgouez

C, Sobel M (1999) Rotary Ni–Ti profile systems for preparing

curved canals in resin blocks: influence of operator on

instrument breakage. International Endodontic Journal 32,

436–43.

Park SH, Cho KM, Kim JW (2003) The efficiency of the Ni–Ti

rotary files in curved simulated canals shaped by novice

operators. Journal of Korean Academy of Conservative Dentistry

28, 146–55.

Peters OA, Peters CI, Schonenberger K, Barbakow F (2003)

ProTaper rotary root canal preparation: assessment of

torque and force in relation to canal anatomy. International


Sattapan B, Nervo GJ, Palamara JE, Messer HH (2000) Defects

in rotary nickel–titanium files after clinical use. Journal of


Schafer E (2001) Shaping ability of Hero 642 rotary nickel–

titanium instruments and stainless steel hand K-Flexofiles in

simulated curved root canals. Oral Surgery Oral Medicine

Oral Pathology Oral Radiology and Endodontics 92, 215–20.

Schafer E, Dzepina A, Danesh G (2003) Bending properties of

rotary nickel–titanium instruments. Oral Surgery Oral Med-

icine Oral Pathology Oral Radiology and Endodontics 96, 757–

63.

Schafer E, Schulz-Bongert U, Tulus G (2004) Comparison of

hand stainless steel and nickel titanium rotary instrumen-

tation: a clinical study. Journal of Endodontics 30, 432–5.

Shen Y, Cheung GS, Bian Z, Peng B (2006) Comparison of

defects in ProFile and ProTaper systems after clinical use.


Timoshenko SP, Goodier JN (1970) Theory of Elasticity. New

York: McGraw-Hill.

Tripi TR, Bonaccorso A, Condorelli GG (2006) Cyclic fatigue of

different nickel–titanium endodontic rotary instruments.

Oral Surgery Oral Medicine Oral Pathology Oral Radiology and

Endodontics 102, e106–14.

Turpin YL, Chagneau F, Vulcain JM (2000) Impact of two

theoretical cross-sections on torsional and bending stresses

of nickel–titanium root canal instrument models. Journal of


Turpin YL, Chagneau F, Bartier O, Cathelineau G, Vulcain JM

(2001) Impact of torsional and bending inertia on root

canal instruments. Journal of Endodontics 27, 333–6.

Walia H, Brantley WA, Gerstein H (1988) An initial investi-

gation of the bending and torsional properties of Nitinol root

canal files. Journal of Endodontics 14, 346–51.

Walsch H (2004) The hybrid concept of nickel–titanium

rotary instrumentation. Dental Clinics of North America 48,

183–202.

Wang GZ (2007) A finite element analysis of evolution of

stress–strain and martensite transformation in front of a

notch in shape memory alloy NiTi. Materials Science and

Engineering A 460–461, 383–91.

Wei X, Ling J, Jiang J, Huang X, Liu L (2007) Modes of failure

of ProTaper nickel–titanium rotary instruments after

clinical use. Journal of Endodontics 33, 276–9.

Xu X, Eng M, Zheng Y, Eng D (2006) Comparative study of

torsional and bending properties for six models of nickel–

titanium root canal instruments with different cross-sec-

tions. Journal of Endodontics 32, 372–5.

Yared GM, Bou Dagher FE, Machtou P (2001) Influence of

rotational speed, torque and operator’s proficiency on

ProFile failures. International Endodontic Journal 34, 47–53.

Yun HH, Kim SK (2003) A comparison of the shaping abilities

of 4 nickel–titanium rotary instruments in simulated root

canals. Oral Surgery Oral Medicine Oral Pathology Oral

Radiology and Endodontics 95, 228–33.



Ex vivo study on root canal instrumentation of tworotary nickel–titanium systems in comparison tostainless steel hand instruments

J. Vaudt1, K. Bitter1, K. Neumann2 & A. M. Kielbassa1

1Department of Operative Dentistry and Periodontology, University School of Dental Medicine, ChariteCentrum 3; and 2Institute

for Biometry and Clinical Epidemiology, ChariteCentrum 4; Charite – Universitatsmedizin Berlin, Berlin, Germany

Abstract

Vaudt J, Bitter K, Neumann K, Kielbassa AM. Ex vivo

study on root canal instrumentation of two rotary nickel–

titanium systems in comparison to stainless steel hand instru-

ments. International Endodontic Journal, 42, 22–33, 2009.

Aim To investigate instrumentation time, working

safety and the shaping ability of two rotary nickel–

titanium (NiTi) systems (Alpha System and ProTaper

Universal) in comparison to stainless steel hand

instruments.

Methodology A total of 45 mesial root canals of

extracted human mandibular molars were selected. On

the basis of the degree of curvature the matched teeth

were allocated randomly into three groups of 15 teeth

each. In group 1 root canals were prepared to size 30

using a standardized manual preparation technique; in

group 2 and 3 rotary NiTi instruments were used

following the manufacturers’ instructions. Instrumen-

tation time and procedural errors were recorded. With

the aid of pre- and postoperative radiographs, apical

straightening of the canal curvature was determined.

Photographs of the coronal, middle and apical cross-

sections of the pre- and postoperative canals were

taken, and superimposed using a standard software.

Based on these composite images the portion of

uninstrumented canal walls was evaluated.

Results Active instrumentation time of the Alpha

System was significantly reduced compared with Pro-

Taper Universal and hand instrumentation (P < 0.05;

anova). No instrument fractures occurred in any of the

groups. The Alpha System revealed significantly less

apical straightening compared with the other instru-

ments (P < 0.05; Mann–Whitney U test). In the apical

cross-sections Alpha System resulted in significantly

less uninstrumented canal walls compared with stain-

less steel files (P < 0.05; chi-squared test).

Conclusion Despite the demonstrated differences

between the systems, an apical straightening effect

could not be prevented; areas of uninstrumented root

canal wall were left in all regions using the various

systems.

Keywords: automated root canal preparation, NiTi

instruments, root canal aberration, root canal shaping,

working safety, working time.

Received 11 April 2008; accepted 16 September 2008

Introduction

The shaping ability of root canal instruments is often

assessed in terms of the preservation of the original root

canal curvature, and without creating iatrogenic

events such as instrument fracture, external transpor-

tation, ledges, or perforations (Weine et al. 1975,

1976). Good canal shaping through mechanical

instrumentation is generally considered essential be-

cause root canal shape may have an effect on the

efficacy of chemical disinfection.

In the last decade, several rotary nickel–titanium

(NiTi) instruments with different configurations and

designs have been developed with the aim to reduce the

preparation time and to simplify the preparation

procedure. Many of these systems have been investi-

gated with regard to their shaping and cleaning ability,

Correspondence: Juliane Vaudt, Abteilung fur Zah-

nerhaltungskunde und Parodontologie, ChariteCentrum 3

fur Zahn-, Mund- und Kieferheilkunde, Charite – Universitats-

medizin Berlin, Aßmannshauser Straße 4-6, D-14197 Berlin,

Germany (Tel.: +49-30-450 562 335 (332); fax: +49-30-450

562 932; e-mail: [email protected]).

doi:10.1111/j.1365-2591.2008.01489.x


handling safety, and working time (Guelzow et al.

2005, Schirrmeister et al. 2006, Sonntag et al. 2007).

These studies have shown that NiTi instruments can

effectively prepare continuously tapered and centred

root canal forms exhibiting only minor deviations from

the main axis of the root canal (Paque et al. 2005,

Schafer et al. 2006, Sonntag et al. 2007). Moreover,

investigations have demonstrated that the use of NiTi

instruments decreased the prevalence and degree of

root canal transportation compared with hand instru-

ments (Schafer & Lohmann 2002, Schafer et al. 2004).

Nevertheless, these effects could not be entirely elim-

inated, and statistically significant differences concern-

ing the straightening effect between rotary NiTi

instruments have been reported (Yun & Kim 2003,

Yoshimine et al. 2005, Schafer et al. 2006). In most

studies the straightening effect has been analyzed

radiographically in the bucco-lingual direction only

(Guelzow et al. 2005, Paque et al. 2005, Schafer et al.

2006). Limited data exist about three-dimensional

morphological changes during preparation.

Total postoperative cleanliness can only be evaluated

histological or by using SEM techniques on longitudinal

or horizontal sections of extracted teeth (Hulsmann

et al. 2005). In addition, the pre- and post-instrumen-

ted root canal cross sections can be analyzed with

respect to unprepared canal walls, thus allowing for

conclusions regarding the mechanical cleaning ability.

The available literature reveals that rotary NiTi instru-

ments shape the coronal and middle third of the root

canal effectively, and create a smooth surface profile

(Foschi et al. 2004, Prati et al. 2004). It has been

reported that the apical third is the critical area of the

root canal, and remaining pulpal and inorganic debris

have been detected (Foschi et al. 2004, Prati et al.

2004). Interestingly, stainless steel hand instruments

revealed equal or even better results concerning

cleaning effectiveness when compared with NiTi

instruments (Schafer & Lohmann 2002, Prati et al.

2004). Various studies reported untreated root canal

wall areas after preparation using rotary NiTi instru-

ments (Peters et al. 2001, 2003). Root canal cleanli-

ness is also dependent on the size of the root canal

instrument. The available literature has revealed that

the use of larger apical instruments led to an advan-

tageous cleaning effect compared with smaller apical

files (Wu &Wesselink 1995, Bartha et al. 2006, Weiger

et al. 2006). However, root canal preparation with

large size instruments can weaken the root, and does

increase the risk of apical transportation (Wu et al.

2003).

In 2005, a newly developed rotary NiTi system

(Alpha System; Brasseler, Lemgo, Germany), and in

2006, an advanced rotary NiTi system (ProTaper

Universal; Dentsply Maillefer, Ballaigues, Switzerland)

were introduced into the market; only limited data exist

on the performance of these systems.

The Alpha System consists of three different instru-

ment sequences according to root canal anatomy

(small, average and large canals). The basic set consists

of five instruments with descending tapers ranging

from 10% to 2%, and sizes from 20 to 35. The

instruments are provided with a titanium nitride

coating, have a five-edged (pentagon) cross-section as

well as a noncutting safety tip. For coronal flaring, an

instrument with an increased taper, a square cross-

section (kite-shaped) and large chip spaces is available

(AF10; access reamer).

ProTaper Universal represents an advancement of

ProTaper, which has been previously investigated in

several studies (Calberson et al. 2004, Guelzow et al.

2005, Paque et al. 2005, Sonntag et al. 2007). The

basic sequence of ProTaper Universal exhibits an

advanced flute design that combines multiple tapers

within the shaft, a convex triangular cross-sectional

design, blades close to the noncutting pilot tip as well as

an increasing chip space (space for the accumulation of

debris) from tip to shaft. A new design feature of

ProTaper Universal NiTi system comprises the more

rounded tips of the finishing files with the aim to

increase the working safety as well as to improve

shaping ability. Furthermore, the cross-section design

has been modified. The convex lateral surfaces of F3 to

F5 are machined to increase its inherent flexibility.

The aim of the present study was to investigate the

instrumentation time, the working safety and the

shaping ability in extracted mandibular molar teeth

with curved root canals using the rotary NiTi systems

Alpha System and ProTaper Universal in comparison to

stainless steel hand instruments. A modified muffle

system was used to enable evaluation both in the

bucco-lingual and in the mesio-distal direction. The

hypothesis was that the parameters instrumentation

time, working safety and shaping ability would be

influenced by the instrumentation technique.


Selection of teeth and experimental set-up

A total of 45 human mandibular molars (extracted for

periodontal reasons) with intact crowns and curved

Vaudt et al. Efficacy of two rotary NiTi instruments


mesial roots were selected. The study protocol con-

formed to the principles outlined in the Central German

Ethics Committee’s statement (Zentrale Ethikkommis-

sion 2007) focusing on the use of human body material

in medical research. Only teeth with completed root

formation, intact root apices, and without visible apical

resorption were included. Coronal access was achieved

using diamond burs, and the mesial root canals were

controlled for apical patency with a size 10 reamer

(VDW, Munich, Germany).

For evaluation of the several parameters a modifica-

tion of the muffle-block, as previously described by

Bramante et al. (1987), Campos & del Rio (1990) and

Hulsmann et al. (1999) was used. A muffle-block was

constructed, which allows removal and exact reposi-

tioning of the complete specimen or sectioned parts of

the latter (Fig. 1). Using this modified muffle, the

exposure of radiographs under reproducible conditions

in two directions (bucco-lingual and mesio-distal) was

guaranteed to take radiographs before, during and after

root canal preparation.

The muffle consisted of a ground section, four lateral

walls and a cover with eight vertical screws. The holder

for the radiographic sensor (fabricated from clear epoxy

resin) could be adjusted at the ground section of the

muffle, and the positioner for the X-ray tube could be

fixed at the outside of the ground part of the muffle

(Fig. 1).

Specimen preparation

The teeth were embedded into the muffle-block with

acrylic resin (Technovit 4071; Heraeus Kulzer, Hanau,

Germany), and shortened coronally to a length of

19 mm. Subsequently, specimens were sectioned hor-

izontally at 3, 6, and 9 mm from the apex (Sagemi-

krotom Leitz 1600; Leica Microsystem, Wetzlar,

Germany). The horizontal segments were remounted

into the muffle, and loss of material (300 lm, because

of the thickness of the saw blade as well as the inter-

slice thickness) was compensated using metal disks of

the corresponding height (300 lm).

An initial size 10 root canal instrument was inserted

into the curved root canal. Standardized radiographs

were taken prior to the instrumentation in a bucco-

lingual as well as in a mesio-distal direction with a

digital radiographic system (Planmeca intra; Planmeca,

Hamburg, Germany), operating at 70 kV and 7 mA.

Thus, the straightening of the instrumented root canals

could be evaluated from two directions to describe the

three-dimensional morphological changes during

preparation.

Root canal curvatures were measured according to a

modified method (Hulsmann & Stryga 1993) using the

software AxioVision (Carl Zeiss MicroImaging; Jena,

Germany). The specimens were divided into three

groups according to the root canal curvature (<25�,25�–35�, and >35�). On the basis of the degree of the

bucco-lingual curvature the teeth were randomized

equally into three groups of 15 teeth each (stratified

random sampling); thus, groups that were equal on the

matching variable were created.

With the aim to analyze the shaping ability and to

draw conclusions regarding the mechanical cleaning

ability of the systems photographs of the preoperative

coronal, middle, and apical cross-sections of the root

canals were taken using a stereo microscope (Stemi

SV11; Carl Zeiss, Oberkochen, Germany) including a

video camera attachment (JVC TK 1070E; Carl Zeiss) at

70· magnification.

Instruments and preparation technique

According to the manufacturers’ instructions all root

canals were initially prepared using a size 10 reamer

followed by a size 15 reamer. Alpha System files were

set into permanent rotation with a 4-level torque limit

setting ENDOadvance handpiece (KaVo; Biberach,

Germany), powered by an electric motor (MF-Perfecta;

W&H, Buermoos, Austria), and using a working speed

Figure 1 Insight into the modified muffle system with fixed

positioner for the X-ray tube (p) and inserted sectioned parts of

specimen (t), mounted for radiographic evaluation in vestib-

ulo-oral direction. The holder (h) bearing the slot for the

radiographic sensor (s) is adjusted at the ground section of the

muffle. The inserted small photograph reveals the tooth spec-

imen rotated at an angle of 90� to enable radiographic

evaluation in mesio-distal direction (for presentation, front

and side sections of the muffle-block have been removed).

Efficacy of two rotary NiTi instruments Vaudt et al.


of 250 rpm (500 rpm for AF10). ProTaper Universal

instruments were set into permanent rotation with

a 4 : 1 reduction handpiece (WD-66 EM; W&H,

Buermoos) powered by a low torque-limited electric

motor (Endo IT control motor; VDW), and a working

speed of 300 rpm was used. Instrumental sequences

followed the manufacturers’ instructions (see Table 1).

The instruments were kept rotating inside the root

canal until they reached the working length and the

instruments designed for crown down technique

(AF10.045, AF06.025, AF04.025) were left in

the root canal for a short period of time (5–8 s).

The manual preparation technique was performed

using stainless steel K-Reamers (VDW) as well as

Hedstrom files (VDW). All root canals were preflared in

the coronal section with number 1 through three

Gates Glidden burs. The K-Reamers were used in a

reaming motion and manipulated in a clockwise

rotation of about 90–120� with a very light inward

force until the file reached the full working distance,

followed by a straight outward pull (turn-and-pull

motion). Hedstrom files were used additionally with

a withdrawing filing motion only. All instruments

were pre-curved. No step-back method of instru-

ment manipulation was used with the hand

instruments.

The individual working length for all specimens was

obtained by measuring the length of the initial

instrument (size 10; VDW) at the apical foramen

subtracting 1 mm. All files were used to instrument

only one canal and coated with a lubricant containing

EDTA (FileCare; VDW) before use. The root canals of all

teeth were instrumented up to size 30. After each

instrument, the root canal was irrigated with 2 mL of

1% NaOCl solution using a syringe and a 28-gauge

needle. All canals were instrumented and analyzed by

the same experienced operator.

Assessment of root canal preparation

Instrumentation time

The active time for root canal instrumentation was

recorded using a digital stopwatch (http://www.jumk.

de/stoppuhr; Internetservice Kummer + Oster, Buchen-

berg, Germany). Time for instrument changes as well as

irrigation times were not considered.

Working safety

The number of fractured instruments during instru-

mentation was documented.

Shaping ability

The assessment of the apical straightening effect for

each system was carried out after preparation up to

sizes 25 and 30, respectively. Radiographs were taken

with a size 25 and 30 instrument from both directions

Table 1 Total number of instruments used, sequence of preparation and working length (WL)

System Sequence

Alpha System (Brasseler, Lemgo, Germany;

five instruments)

Average canals (as indicated by the manufacturer)

AF10.045 – 10% taper, size 45 (canal orifice)

AF06.025 – 6% taper, size 25 (crown down to the curvature)

AF04.025 – 4% taper, size 25 (crown down to the curvature)

AF02.025 – 2% taper, size 25 (WL)

AF02.030 – 2% taper, size 30 (WL)

ProTaper Universal (Dentsply Maillefer,

Ballaigues, Switzerland; seven instruments)

S1 file (shaping file 1) - 2–11% taper, size 17 (canal orifice)

SX (auxiliary shaper file) - 3–19% taper, size 19 (canal orifice)

S1 (shaping file 1) - 2–11% taper, size 17 (WL)

S2 (shaping file 2) - 4–11.5% taper, size 20 (WL)

F1 (finishing file 1) - 7–5.5% taper, size 20 (WL)



Stainless steel instruments (Vereinigte

Dentalwerke, Germany; eight instruments)

Reamer - 2% taper, size 15 (WL)

Hedstrom file - 2% taper, size 15 (WL)









(bucco-lingual and mesio-distal) (Fig. 2). Based on

the canal curvatures assessed pre- and postoperatively,

the apical root canal straightening was determined as

the difference between apical root canal curvatures

before and after instrumentation using instruments of

size 25 and 30, respectively.

In the coronal, middle and apical cross-sections of

the root canal the portion of uninstrumented canal

walls was evaluated. After root canal preparation

postoperative photographs of the cross sections were

taken. Pre- and postoperative photographs were super-

imposed using reference marks (Fig. 3) (Corel Draw;

Corel Corporation, Unterschleißheim, Germany).

On the basis of these images the distance of contact

between the pre- and postoperative root canal walls

was measured, and the portion of uninstrumented root

canal walls was determined (Fig. 3). According to the

results, specimens were divided into four groups:

Group 1: 0–25% contact between pre- and

postoperative cross-section/root canal wall.

Group 2: >25% contact between pre- and






Statistical analyses

The statistical analysis was conducted using one-way

anova with post hoc Tukey B tests for the active

(a) (b)

(c) (d)

Figure 2 Representative example of a series of radiographs

taken in vestibulo-oral (a) and mesio-distal direction before (c)

and after preparation up to ISO size 30 (b, d). The root canal

curvature was measured prior to instrumentation with the

initial instrument inserted (reamer ISO size 10) (a, c). Based on

the radiographs taken after preparation up to ISO size 30 (b,

d), canal curvature could be measured and apical straighten-

ing could be determined.

Figure 3 Representative superimposi-

tion of the pre- and postoperative

photographs of the root canal cross-

sections using reference marks (apical

area). The bolts demonstrate the edge of

the half-transparent superimposed

photograph of the instrumented canal.

The coloured lines show the traced root

canal outlines (red = initial outline;

black = outline after root canal prepa-

ration). Note that all root canal walls

were instrumented in this specimen.



instrumentation time, and Kruskall–Wallis test fol-

lowed by pairwise comparisons using Mann–Whitney

U tests for the straightening effect. The Wilcoxon test

for paired samples was used for comparisons of both

directions (bucco-lingual and mesio-distal) regarding

the degree of straightening. The percentages of unin-

strumented root canal walls were compared using chi-

squared test (exact test was used). The level of

significance was set at a = 0.05 (without adjusting

for the respective comparisons, as the described pro-

ceeding is equivalent to the closed test procedure for the

particular case of three study groups). All statistical

analyses were performed using spss version 15.0

software (SPSS; Chicago, IL, USA).

Results

Working time

The active time for root canal instrumentation

(Table 2) was affected significantly by the systems used

(P < 0.0005; anova). The Alpha System required

significantly less time compared with the other systems

(P < 0.05; Tukey B), whereas ProTaper Universal

revealed a significantly reduced instrumentation time

compared with the manual technique (P < 0.05;

Tukey B).

Working safety

During the preparation of the curved root canals, no

fractures of any of the used instruments (stainless steel

files as well as the NiTi instruments) could be observed.

Shaping ability

The apical straightening of the curved root canals was

significantly affected by the instrumentation system

(Kruskall–Wallis). Results and P-values (for compari-

sons of all techniques, and with regard to the respective

preparations sizes) are summarized in Table 3. In

general, straightening was more pronounced with size

30 instruments compared with size 25; differences

amongst the groups were increased (as indicated by the

P-values) after use of size 30 instruments. In both

directions (bucco-lingual and mesio-distal), the use of

stainless steel instruments resulted in significantly

increased (P < 0.05; Wilcoxon) straightening if com-

pared with Alpha files; differences between ProTaper

Universal and the manual technique were not signif-

icant. No statistically significant differences between

the two directions regarding the degree of straightening

could be observed.

The analysis of the pre- and postoperative cross-

sections revealed that all systems used left uninstru-

mented root canal walls in all regions. For the coronal

and middle cross-sections no differences between the

systems were found with respect to uninstrumented

root canal walls (P > 0.05; Kruskal–Wallis). The

portion of uninstrumented root canal walls in the

apical cross-sections was significantly affected by the

instrumentation system (P = 0.004; chi-squared test);

the results are summarized in Table 4. Instrumentation

using the Alpha System resulted in significantly less

unprepared root canal walls compared with the man-

ual technique (P = 0.001; chi-squared test). Compar-

ison between ProTaper Universal and stainless steel

files did not reveal any significant differences

(P = 0.153; chi-squared test). ProTaper Universal left

more unprepared root canal walls compared with

Alpha System; however, this statistical difference was

only weakly significant (P = 0.043; chi-squared test).

Discussion

The aim of the present investigation was to compare

the shaping ability of two recently introduced rotary

NiTi instruments in contrast to a manual technique

using stainless steel instruments. The results revealed

significant differences between the used systems with

respect to their shaping ability as well as regarding the

working time. Thus, the hypothesis of the present study

concerning the differences between the systems regard-

ing the evaluated parameters could not be rejected.

Study design

Previous investigations that focused on the shaping

ability of NiTi instruments used either simulated root

canals (Yun & Kim 2003, Yoshimine et al. 2005,

Schirrmeister et al. 2006) or extracted human teeth

(Paque et al. 2005, Schafer et al. 2006). Simulated root

Table 2 Mean active preparation time (in s) and standard

deviation (SD)

System Mean SD

Alpha Systema 103.2 13.5

ProTaper Universalb 150.7 18.9

Manual techniquec 238.3 35.1

a,b,cMeans with differing superscript letters indicate significant

differences at a = 0.05.



canals using pre-fabricated resin blocks allow for

standardization of degree, location and radius of root

canal curvature in three dimensions as well as the root

canal length (Peters 2004, Hulsmann et al. 2005).

However, the hardness and abrasion behaviour of

acrylic resin and root dentine is not identical, and

consequently does not reflect the action of the instru-

ments in root canals of real teeth (Peters 2004,

Hulsmann et al. 2005). Therefore, using extracted

human teeth reflects the clinical situation more

adequately. Nevertheless, large variations concerning

root canal morphology and dentine hardness compli-

cate standardization of groups (Hulsmann et al. 1999).

Similar apical preparation diameters are required for

the comparison of the shaping and cleaning ability of

different root canal instruments. Thus, in all investi-

gated groups of the present investigation, the final

apical preparation diameter was size 30. Moreover, to

reduce the wide range of variations in three-dimen-

sional root canal configuration the present study used

mesial root canals of first and second mandibular

molars. These teeth reveal root canal curvatures in

most cases. Consequently, the measured degrees of the

root canal curvature were categorized into three groups

according to a modified method described by Schneider

(Schneider 1971); this matched-group design allowed

for minimization of high variations in the degree of

curvature between the groups.

To evaluate the quality of root canal preparation a

study design is desirable that allows for standardization

and facilitates simulation of the clinical situation.

Additionally, all relevant parameters should be

recorded. Root canal morphology and the effect of

instrumentation have been studied via numerous

techniques (Campos & del Rio 1990, Hulsmann et al.

1999).

A method has been introduced (Bramante et al.

1987) and modified (Campos & del Rio 1990, Huls-

mann et al. 1999) in which root canals can be

analyzed before and after instrumentation using

extracted teeth. Previously published literature has

described the varying configurations of the used muffle-

blocks (Campos & del Rio 1990, Hulsmann et al. 1999).

Various elements, horizontal and vertical grooves in

the walls of the muffle-blocks have been designed and

integrated to guarantee the exact reposition of the

specimen. In the present investigation a muffle-block

was constructed to evaluate simultaneously both

mechanical cleaning and shaping ability under

Table 3 Mean values (including standard deviations as well as minimum and maximum values) of apical straightening in

both directions (in degrees)

System

Straightening of vestibulo-oral direction (in degrees)

ISO size 25 ISO size 30

Mean SD Min Max Mean SD Min Max

Alpha System 0.6a 0.5 0.0 1.4 1.4a 1.0 0.2 3.8

ProTaper Universal 2.7a,b 2.8 0.0 9.2 4.3b 3.4 0.1 12.5

Manual technique 2.9b 3.0 0.1 9.4 4.4b 3.7 0.5 13.2

P-value (Kruskal-Wallis) 0.043 0.012

System

Straightening of mesio-distal direction (in degrees)

ISO size 25 ISO size 30

Mean SD Min Max Mean SD Min Max

Alpha System 1.7a 1.6 0.0 6.4 2.4a 1.9 0.1 7.0

ProTaper Universal 3.4b 2.2 0.0 7.7 4.4b 2.1 0.2 7.8

Manual technique 3.5b 2.9 0.1 11.7 4.7b 3.8 0.8 16.4

P-value (Kruskal-Wallis) 0.028 0.019

a,b,cSignificant differences (P < 0.05) according to comparisons using Mann–Whitney U test are indicated by different superscript

letters.

Table 4 Portion of uninstrumented area between superim-

posed pre- and postoperative root canal walls in the apical

region when viewed in cross-sections

System

Contact between pre- and

postoperative cross-section (%)

0–25% >25% >50% >75%

Alpha System Number 14 1 0 0

percentage 93.3% 6.7% 0.0% 0.0%

ProTaper

Universal

Number 8 5 2 0

percentage 53.3% 33.3% 13.3% 0.0%

Manual

technique

Number 3 10 1 1

percentage 20.0% 66.6% 6.7% 6.7%



reproducible conditions. The elements were designed to

facilitate removal and exact repositioning of the com-

plete specimen or sectioned parts of the latter. On the

basis of the tapered internal space, integrated position-

ing devices as well as a cover, the exact and nonrelo-

catable vertical and horizontal position of the specimen

was guaranteed.

After sectioning of the embedded teeth horizontally,

the specimens were remounted into the muffle for

instrumentation. The loss of material was predictable

because of the use of the saw microtome, and could be

compensated using spacers (metal disks) of the same

height. The configuration of the used muffle system

allowed for evaluation of the root canal cross-section

prior, during and after instrumentation without chang-

ing the three-dimensional morphology of root canal.

A further important improvement was the ability to

evaluate the parameter ‘straightening of curved root

canals’ in two directions (bucco-lingual and mesio-

distal) with the aim to describe the three-dimensional

morphological changes during preparation. This

parameter refers to the maintenance of the original

shape of curved root canals, and provides information

about the direction of removed material.

Shaping ability

This study showed that root canal preparation using

stainless steel instruments as well as NiTi systems

results in a pronounced apical straightening effect in

the bucco-lingual as well as in the mesio-distal plane.

No differences were found between the two directions

regarding the degree of root canal transportation.

Various investigations demonstrated that the use of

NiTi instruments decreased the prevalence and degree

of root canal transportation compared with stainless

steel hand instruments (Schafer & Lohmann 2002,

Schafer et al. 2004). However, other studies reported

no differences between rotary NiTi systems and stain-

less steel hand instruments regarding root canal

transportation (Guelzow et al. 2005, Hartmann et al.

2007). These divergent outcomes might be explained

by differences in methodologies, methods of assessment,

instruments and preparation techniques.

A further aspect is the design of an instrument that

might influence the shaping ability. Stainless steel files

are relatively stiff that will increase with larger instru-

ment size and causes high lateral forces in curved root

canals (Bergmans et al. 2001, Schafer & Tepel 2001).

The rigidity of an instrument could be responsible for

straightening of and aberration from the root canal

(including ledges, zipping and perforations), along with

leaving significant portions of the root canal wall

uninstrumented (Peters et al. 2003, Calberson et al.

2004). It has been assumed that NiTi instruments

could improve shaping ability and minimize any

aberrations during root canal preparation (Paque et al.

2005, Yoshimine et al. 2005, Schirrmeister et al. 2006,

Sonntag et al. 2007). However, these effects could not

be entirely eliminated, and differences amongst rotary

NiTi instruments have been demonstrated (Yun & Kim

2003, Yoshimine et al. 2005, Schafer et al. 2006).

With regard to the reported outcomes, it has to be

stressed that different rake angles of instruments should

reveal varying cutting efficacies; indeed, most rotary

files have negative rake angles with a predominantly

scraping motion. In general, the shaping ability of root

canal instruments is a complex interrelationship of

various parameters such as cross-sectional design, chip

removal capacity, helical and rake angle, metallurgical

properties and surface treatment of the instrument

(Schafer 1999, Schafer & Oitzinger 2008).

The more rounded tip of the finishing files in the

ProTaper Universal sequence has been developed to

increase the working safety as well as to improve the

shaping ability. Furthermore, the cross-section design

has been modified to increase its inherent flexibility.

Obviously, the advanced design features of the ProTa-

per Universal system revealed similar results compared

with previous studies evaluating the classic ProTaper

files (Peters et al. 2003, Calberson et al. 2004, Sonntag

et al. 2007).

This study found a significantly pronounced apical

straightening effect in both directions using ProTaper

Universal compared to Alpha System. The different

instrument designs of these NiTi systems (i.e. progres-

sive versus constant taper) could have influenced the

observed outcomes. Previously published studies dem-

onstrated relationships between bending moment and

cross-section, file size as well as taper of an instrument

(Schafer & Tepel 2001, Schafer et al. 2003). The

ProTaper Universal files have multiple and increased

tapers within the shaft compared with the Alpha files

presenting a less and constant taper. The apical

transportation towards the outer aspects of the root

canal could have been affected by the variable tapers

along the cutting surface of the ProTaper Universal

files (up to 11% at the tip) compared with the

moderately tapered (2%) Alpha System instruments.

An increasing taper is associated with increased cross-

section areas and, accordingly, with decreased flexi-

bility of the files that could cause straightening and



root canal aberration during preparation (Bergmans

et al. 2001). The decreasing taper sequence of the

finishing files enhance the strength of the instruments,

but increase the stiffness of their tips (i.e. ProTaper

Universal size 30 is 9%, size 20 is 7%), thus resulting

in high lateral forces.

Previously published data about classic ProTaper

files showed similar results compared with the present

investigation and demonstrated varying degrees of root

canal straightening and transportation (Peters et al.

2003, Yoshimine et al. 2005). These tended at the mid-

point of the curvature towards the inner aspects and

apically towards the outer aspects of the root canal

(Calberson et al. 2004, Sonntag et al. 2007). An

investigation comparing three rotary NiTi instruments

has demonstrated a tendency to ledge or zip formation

at the end-point of preparation using ProTaper com-

pared with RaCe and K3 (Yoshimine et al. 2005). The

RaCe and K3 groups showed favourable results, and

the prepared root canals displayed a smooth shift to the

original root canals at the end point. The authors

recommended that more flexible instruments, like K3

and RaCe, should be used in the apical preparation of

curved root canals (Yoshimine et al. 2005). Compari-

sons amongst different rotary NiTi systems instruments

revealed more root canal straightening after use of

ProTaper compared to Mtwo, K3, ProFile, GT Rotary,

and Quantec (Yun & Kim 2003, Sonntag et al. 2007).

In contrast to these observations, other investigations

found no statistical differences in root canal transpor-

tation using ProTaper compared with other rotary NiTi

instruments after preparation of up to size 30 (Guelzow

et al. 2005, Paque et al. 2005).

The rotary NiTi Alpha System files provided a

centred apical preparation and maintained the original

shape of the curved root canals with only minor

deviation from the main axis. The files are character-

ized by a pentagon-type cross-sectional design resulting

in only slightly positive cutting angles and a compa-

rable low chip space (Schafer & Oitzinger 2008). The

reduced root canal transportation could be explained

by the high flexibility of these instruments because of

their minor and constant taper along the cutting

surface. This superior flexibility reduces the risk of root

canal transportation during the enlargement of curved

root canals (Schafer et al. 2003).

Furthermore, it has been shown that the cross-

sectional design as well as cross-sectional areas

mainly influenced bending properties of instruments

(Schafer & Tepel 2001, Schafer et al. 2003). It has

been reported that for identical working diameters, the

area of a triple-helix cross-section was found to be

approximately 30% greater than that of triple-U file

(Turpin et al. 2000). Because of the more massive

structure of a triple helix file, this instrument was

found to be less flexible than the triple-U instrument

(Turpin et al. 2000). The Alpha files used in the

present study had a pentagon-type cross-section. It

might be speculated that this form of cross-section

results in a high core diameter and a high cross-

sectional area compared with other forms (i.e. trian-

gular, square cross-section), and, consequently, in

reduced flexibility. Furthermore, the small chip space

could lead to apical blockage caused by insufficient

transportation of debris towards the orifice (Bergmans

et al. 2001). Alpha system files are the only known

rotary NiTi systems with a pentagon cross-sectional

design, and no published literature exists about this

design feature up to now.

Obviously, in the case of comparison between Alpha

System and ProTaper Universal the taper had a greater

influence on the flexibility than the cross sectional

design. Under the conditions of the present study the

use of the rotary NiTi Alpha System files resulted in

minimal apical root canal transportation. However, the

influence of the individual geometric characteristics of

the instruments on the cleaning and shaping ability

remains speculative.

It is well known that the surface hardness of NiTi

instruments is lower than that of stainless steel

instruments (Brockhurst & Hsu 1998, Schafer &

Oitzinger 2008). Consequently, the cutting efficiency

should be less compared with most stainless steel

instruments (Brockhurst & Hsu 1998). With the aim

to improve the surface hardness (and thereby increas-

ing the shaping efficiency of NiTi instruments) several

surface engineering techniques have been used, i.e.

physical vapour deposition (PVD) techniques. Recent

studies have shown that the PVD technique is suitable

to significantly increase the cutting efficiency of NiTi

instruments (Schafer 2002). However, those findings

did not corroborate the observation of a previously

published study comparing the cutting efficiency of

different NiTi systems (Schafer & Oitzinger 2008);

here, the results revealed no significant influence of

the PVD coating surface on the cutting efficiency, and

the Alpha System files showed a significantly lower

cutting efficiency compared with Mtwo, RaCe, and

Flexmaster (Schafer & Oitzinger 2008). Thus, the

influence of the PVD coating on the cutting efficacy of

NiTi instruments with different cross-sectional design

remains unclear.



Unprepared root canal wall areas

The comparison of the pre- and postoperative photo-

graphs of root canal cross-sections enables conclusions

on shaping ability as well as mechanical cleaning

ability. The prepared root canal should include the

original root canal dimensions, and no unprepared

areas should remain (compare Fig. 3). In the present

investigation analysis of the pre- and postoperative

cross-sections showed that the manual technique using

stainless steel instruments as well as the rotary NiTi

systems left uninstrumented root canal walls in all

regions. While the coronal and middle cross-sections

demonstrated sufficient shaping outcomes with only

minor untreated areas, the percentage of uninstru-

mented root canal walls in the apical cross-section was

significantly affected by the instrumentation system.

Following preparation using the Alpha System the

specimens showed the lowest percentage of unprepared

root canal outlines compared with the manual tech-

nique using stainless steel instruments and the ProTa-

per Universal that ranged between the two previous

groups.

These findings corroborate the results of previous

studies. It has been reported that in the apical part of

the root canal amounts of remaining pulpal and

inorganic debris could be detected after using rotary

NiTi instruments (Foschi et al. 2004, Prati et al. 2004).

Furthermore, results indicated large untreated areas of

the root canal walls (Peters et al. 2001, 2003). These

areas tended to be on the convex curvature at mid-root

and the concave side of the curvature more apically

(Peters et al. 2003). Evaluation of the original ProTaper

resulted in untouched areas ranging from 43% to 49%

(Peters et al. 2003).

The superior shaping efficiency of the Alpha System

could be attributed to the high elasticity characteristics

of the instruments that resulted in minor root canal

transportation during root canal preparation. Conse-

quently, the files remove dentine uniformly on the

outer and inner side of the root canal and only minor

areas remain untouched. However, further studies

should evaluate whether the preparation of wide root

canals using the less tapered Alpha files will result in

sufficient cleanliness.

Nevertheless, the clinical significance of the param-

eter ‘prepared surface’ is not yet clarified. However,

when considering the fact that viable microbes may

penetrate deep into the dentinal tubules and may

persist during root canal treatment (Chuste-Guillot

et al. 2006), the need of an efficient irrigation (in

addition to the shaping regime) to clean the root canals

effectively is clearly highlighted.

Instrumentation time

The evaluation of the parameter ‘working time’ should

demonstrate the efficacy of a system and its clinical

suitability. Studies that investigated the working time

of various NiTi systems used the latter in different

treatment sequences and changing number of files.

Notwithstanding, working time depends on preparation

technique and operator experience.

Some investigations evaluated working time as the

active instrumentation time (summation of time taken

for files to work inside the root canal) (Paque et al.

2005). Other studies measured the working time

including the active instrumentation time as well as

the time for changing instruments and irrigation, thus

resulting in considerably higher values (Schirrmeister

et al. 2006). Evaluations of manual techniques and

rotary NiTi instruments have demonstrated huge

variations of working time, and cannot provide recom-

mendations for one of the two techniques (Schafer et al.

2004, Guelzow et al. 2005, Schirrmeister et al. 2006).

The present investigation observed shorter working

times for NiTi preparation compared with the manual

instrumentation. The results indicate that the ProTaper

Universal system required more time to prepare a root

canal than the Alpha System.

These results may be explained by the varying

number of instruments. In the present investigation,

root canal preparation was performed using eight

stainless steel files for the manual technique, and seven

files for ProTaper Universal. The five Alpha System files

exhibited the lowest number of required instruments

for root canal enlargement.

Working safety

The reasons for fractures of rotary NiTi instruments are

multifactorial, and complications can be attributed to

instrument design, manufacturing process, canal con-

figuration, applied force during instrumentation, prep-

aration technique, operator’s skills and experience as

well as the number of application inside the root canal

(Parashos & Messer 2006). In the present investigation

no fractures of the stainless steel files as well as the NiTi

instruments could be observed. All instruments were

used to instrument only one single root canal. It should

be emphasized that the regimen used was owing to the

objectives (to compare the cleaning and shaping ability



of different systems). Thus, this procedure does not

adequately reflect the clinical situation, and the clinical

relevance concerning the working safety should be

interpreted with caution.

Conclusion

Within the limitations of the present ex vivo study, the

experimental results suggest that none of the rotary

NiTi systems used was able to impede an apical

straightening effect during root canal preparation;

uninstrumented root canal wall areas were left in all

regions with all systems.

Few statistically significant differences amongst the

three instrumentation techniques could be revealed.

Instrumentation of curved root canals to ISO size 30

using instruments with greater taper (Pro Taper

Universal F2, F3) and stiffer instruments (Pro Taper

Universal F3, stainless steel file of size 30) seemed to

result in increased root canal transportation and in a

higher portion of unprepared root canal walls com-

pared with flexible NiTi instruments (Alpha 02/30)

that maintained the curvature of the root canal.

Acknowledgements

The authors are indebted to Brasseler (Lemgo,

Germany), Dentsply Maillefer (Ballaigues, Switzerland)

and Vereinigte Dentalwerke (Munich, Germany) for

generously providing the instruments.

References

Bartha T, Kalwitzki M, Lost C, Weiger R (2006) Extended

apical enlargement with hand files versus rotary NiTi files.

Part II. Oral surgery, Oral medicine, Oral Pathology, Oral

Radiology, and Endodontics 102, 692–7.

Bergmans L, Van Cleynenbreugel J, Wevers M, Lambrechts P

(2001) Mechanical root canal preparation with NiTi rotary

instruments: rationale, performance and safety. Status

report for the American Journal of Dentistry. American

Journal of Dentistry 14, 324–33.

Bramante CM, Berbert A, Borges RP (1987) A methodology

for evaluation of root canal instrumentation. Journal of


Brockhurst P, Hsu E (1998) Hardness and strength of

endodontic instruments made from NiTi alloy. Australian


Calberson FL, Deroose CA, Hommez GM, De Moor RJ (2004)

Shaping ability of ProTaper nickel-titanium files in simu-

lated resin root canals. International Endodontic Journal 37,

613–23.

Campos JM, del Rio C (1990) Comparison of mechanical and

standard hand instrumentation techniques in curved root

canals. Journal of Endodontics 16, 230–4.

Chuste-Guillot MP, Badet C, Peli JF, Perez F (2006) Effect of

three nickel-titanium rotary file techniques on infected root

dentin reduction. Oral Surgery, Oral Medicine, Oral Pathology,

Oral Radiology, and Endodontics 102, 254–8.

Foschi F, Nucci C, Montebugnoli L et al. (2004) SEM evalu-

ation of canal wall dentine following use of Mtwo and

ProTaper NiTi rotary instruments. International Edodontic

Journal 37, 832–9.

Guelzow A, Stamm O, Martus P, Kielbassa AM (2005)

Comparative study of six rotary nickel-titanium systems

and hand instrumentation for root canal preparation.


Hartmann MS, Barletta FB, Camargo Fontanella VR, Vanni JR

(2007) Canal transportation after root canal instrumenta-

tion: a comparative study with computed tomography.


Hulsmann M, Stryga F (1993) Comparison of root canal

preparation using different automated devices and hand

instrumentation. Journal of Endodontics 19, 141–5.

Hulsmann M, Gambal A, Bahr R (1999) An improved

technique for the evaluation of root canal preparation.


Hulsmann M, Peters OA, Dummer PMH (2005) Mechanical

preparation of root canals: shaping goals, techniques and

means. Endodontic Topics 10, 30–76.

Paque F, Musch U, Hulsmann M (2005) Comparison of root

canal preparation using RaCe and ProTaper rotary Ni-Ti

instruments. International Endodontic Journal 38, 8–16.

Parashos P, Messer HH (2006) Rotary NiTi instrument

fracture and its consequences. Journal of Endodontics 32,

1031–43.

Peters OA (2004) Current challenges and concepts in the

preparation of root canal systems: a review. Journal of


Peters OA, Schonenberger K, Laib A (2001) Effects of four

Ni-Ti preparation techniques on root canal geometry

assessed by micro computed tomography. International


Peters OA, Peters CI, Schonenberger K, Barbakow F (2003)

ProTaper rotary root canal preparation: effects of canal

anatomy on final shape analysed by micro CT. International


Prati C, Foschi F, Nucci C, Montebugnoli L, Marchionni S

(2004) Appearance of the root canal walls after preparation

with NiTi rotary instruments: a comparative SEM investi-

gation. Clinical Oral Investigations 8, 102–10.

Schafer E (1999) Relationship between design features of

endodontic instruments and their properties. Part 1. Cutting

effiency. Journal of Endodontics 25, 52–5.

Schafer E (2002) Effect of physical vapor deposition on cutting

efficiency of nickel-titanium files. Journal of Endodontics 28,

800–2.



Schafer E, Lohmann D (2002) Efficiency of rotary nickel-

titanium FlexMaster instruments compared with stainless

steel hand K-Flexofile. Part 2. Cleaning effectiveness and

instrumentation results in severely curved root canals of

extracted teeth. International Endodontic Journal 35, 514–21.

Schafer E, Oitzinger M (2008) Cutting efficiency of five

different types of rotary nickel-titanium instruments. Journal

of Endodontics 34, 198–200.

Schafer E, Tepel J (2001) Relationship between design features

of endodontic instruments and their properties. Part 3.

Resistance to bending and fracture. Journal of Endodontics

27, 299–303.

Schafer E, Dzepina A, Danesh G (2003) Bending properties of

rotary nickel-titanium instruments. Oral Surgery, Oral Med-

icine, Oral Pathology, Oral Radiology, and Endodontics 96,

757–63.

Schafer E, Schulz-Bongert U, Tulus G (2004) Comparison of

hand stainless steel and nickel titanium rotary instrumen-

tation: a clinical study. Journal of Endodontics 30, 432–5.

Schafer E, Erler M, Dammaschke T (2006) Comparative study

on the shaping ability and cleaning efficiency of rotary

Mtwo instruments. Part 2. Cleaning effectiveness and

shaping ability in severely curved root canals of extracted

teeth. International Endodontic Journal 39, 203–12.

Schirrmeister JF, Strohl C, Altenburger MJ, Wrbas KT, Hellwig

E (2006) Shaping ability and safety of five different rotary

nickel-titanium instruments compared with stainless steel

hand instrumentation in simulated curved root canals. Oral

Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and


Schneider SW (1971) A comparison of canal preparations in

straight and curved root canals. Oral Surgery, Oral Medicine,

and Oral Pathology 32, 271–5.

Sonntag D, Ott M, Kook K, Stachniss V (2007) Root canal

preparation with the NiTi systems K3, Mtwo and ProTaper.


Turpin YL, Chagneau F, Vulcain JM (2000) Impact of two

theoretical cross-sections on torsional and bending stresses

of nickel-titanium root canal instrument models. Journal of


Weiger R, Bartha T, Kalwitzki M, Lost C (2006) A clinical

method to determine the optimal apical preparation size.

Part I. Oral Surgery, Oral Medicine, Oral Pathology, Oral


Weine FS, Kelly RF, Lio PJ (1975) The effect of preparation

procedures on original canal shape and on apical foramen

shape. Journal of Endodontics 1, 255–62.

Weine FS, Kelly RF, Bray KE (1976) Effect of preparation with

endodontic handpieces on original canal shape. Jounal of


Wu MK, Wesselink PR (1995) Efficacy of three techniques in

cleaning the apical portion of curved root canals. Oral

Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and


Wu MK, van der Sluis LW, Wesselink PR (2003) The

capability of two hand instrumentation techniques to

remove the inner layer of dentine in oval canals. Interna-


Yoshimine Y, Ono M, Akamine A (2005) The shaping effects

of three nickel-titanium rotary instruments in simulated S-

shaped canals. Journal of Endodontics 31, 373–5.

Yun HH, Kim SK (2003) A comparison of the shaping abilities

of 4 nickel-titanium rotary instruments in simulated root

canals. Oral Surgery, Oral Medicine, Oral Pathology, Oral


Zentrale Ethikkommission (2007) The use of human body

materials for the purposes of medical research. URL http://

www.zentrale-ethikkommission.de/page.asp?his=0.1.21

[accessed on 6th July 2008].



Dislocation resistance of ProRoot Endo Sealer, acalcium silicate-based root canal sealer, fromradicular dentine

B. P. Huffman1, S. Mai2, L. Pinna3, R. N. Weller4, C. M. Primus5, J. L. Gutmann6, D. H. Pashley7

& F. R. Tay4,7

1School of Dentistry, Medical College of Georgia, Augusta, GA, USA; 2Guanghua School of Stomatology & Institute of

Stomatological Research, Sun Yat-sen University, Guangzhou, China; 3Universita degli Studi di Cagliari, Reparto di Odontoiatria

Conservatrice, Sardinia, Italy; 4Department of Endodontics, School of Dentistry, Medical College of Georgia, Augusta, GA, USA;5Primus Consulting, Bradenton, FL, USA; 6Department of Endodontics, Baylor College of Dentistry, Texas A&M University System

Health Science Center, Dallas, TX, USA; and 7Department of Oral Biology, School of Dentistry, Medical College of Georgia, Augusta,

GA, USA

Abstract

Huffman BP, Mai S, Pinna L, Weller RN, Primus CM,

Gutmann JL, Pashley DH, Tay FR. Dislocation resistance

of ProRoot Endo Sealer, a calcium silicate-based root

canal sealer, from radicular dentine. International Endodontic

Journal, 42, 34–46, 2009.

Aim To examine the dislocation resistance of three

root canal sealers from radicular dentine with and

without immersion in a simulated body fluid (SBF),

using a modified push-out test design that produced

simulated canal spaces of uniform dimensions under

identical cleaning and shaping conditions.

Methodology Sixty single-rooted caries-free human

canine teeth were used. Standardized simulated canal

spaces were created using 0.04 taper ProFile instru-

ments along the coronal, middle and apical thirds of

longitudinal tooth slabs. Following NaOCl/ethylenedi-

amine tetra-acetic acid cleaning, the cavities were filled

with ProRoot Endo Sealer, AH Plus Jet or Pulp Canal

Sealer. After setting, half of the cavities were tested

with a fibre-optic light-illuminated push-out testing

device. The rest were immersed in SBF for 4 weeks

before push-out evaluation. Failure modes were exam-

ined with stereomicroscopy and field emission (FE)-

scanning electron microscopy.

Results Location of the sealer-filled cavities did not

affect push-out strengths. ProRoot Endo Sealer exhibited

higher push-out strengths than the other two sealers

particularly after SBF storage (P < 0.001). Failure

modes were predominantly adhesive and mixed for Pulp

Canal Sealer and AH Plus Jet, and predominantly

cohesive for ProRoot Endo Sealer. Spherical amorphous

calciumphosphate-like phases that spontaneously trans-

formed into apatite-like phaseswere seen in the fractured

specimens of ProRoot Endo Sealer after SBF storage.

Conclusions When tested in bulk without a main

core, both ‘sealer type’ and ‘SBF storage’ were signif-

icant in affecting push-out results. The ProRoot Endo

Sealer demonstrated the presence of spherical amor-

phous calcium phosphate-like phases and apatite-like

phases (i.e. ex vivo bioactivity) after SBF storage.

Keywords: calcium silicate-based sealer, dislocation

resistance, in vitro bioactivity, thin-slice push-out test.

Received 1 June 2008; accepted 16 September 2008

Introduction

The use of a sealer and a thermoplastic core material

for filling root canals is the accepted norm in contem-

porary root canal procedures. As leakage from the

apical or coronal direction is a possible cause of root

treatment failure (Madison & Wilcox 1988, De Moor &

Hommez 2000), a root canal sealer should exhibit good

sealing (Laghios et al. 2000) and adhesive properties

(Wennberg & Ørstavik 1990, Gettleman et al. 1991,

Timpawat et al. 2001, Lee et al. 2002a,b, Saleh et al.

2003, Tagger et al. 2003). A sealer may be

Correspondence: Dr Franklin R. Tay, Department of Endodon-

tics, School of Dentistry, Medical College of Georgia, Augusta,

GA, 30912-1129, USA (Tel.: +1 706 7212033; fax:

+1 706 7216252; e-mail: [email protected]).

doi:10.1111/j.1365-2591.2008.01490.x


conceptualized as a joint created between the radicular

dentine and the filling material. Similar to other

prosthetic joints in the body, the ability to resist

dislocation during function is crucial to their survival

(Scifert et al. 1999, Weale et al. 2002, He et al. 2007).

For a root canal sealer, the ability to resist disruption of

the established seal via micromechanical retention or

friction is highly desirable during intraoral tooth

flexure (Panitvisai & Messer 1995) or preparation of

cores or postspaces along the coronal- and middle-

thirds of canal walls (Munoz et al. 2007).

Predictable clinical results have been reported with

the use of gutta-percha in conjunction with zinc oxide

eugenol or epoxy resin-based root canal sealers (Saleh-

rabi & Rotstein 2004, Tilashalski et al. 2004). Never-

theless, there is a continuous quest for alternative

sealers or root filling materials with better seal and dis-

location resistance. Although the correlation between

the sealing property of a root canal sealer and its

adhesive characteristics has not been firmly esta-

blished, it is essential that the dislocation resistance of

a root canal sealer to dentine is not adversely affected

by the seepage of body fluids when there is a breach of

either the apical or coronal seal.

ProRoot Endo Sealer (Dentsply Tulsa Dental Special-

ties, Tulsa, OK, USA) is an experimental calcium

silicate-based root canal sealer that is designed to be

used in conjunction with a root filling material in either

the cold lateral, warm vertical or carrier-based filling

techniques. The major components of the powder

component are tricalcium silicate and dicalcium sili-

cate, with the inclusion of calcium sulphate as a setting

retardant, bismuth oxide as a radiopacifier and a small

amount of tricalcium aluminate. The liquid component

consists of a viscous aqueous solution of a water-

soluble polymer. Similar to other tricalcium silicate and

dicalcium silicate-containing biomaterials, the sealer

produces calcium hydroxide on reaction with water

(Gou et al. 2005, Camilleri & Pitt Ford 2006, Wang

et al. 2008). It is also anticipated that release of

calcium and hydroxyl ions from the set sealer will

result in the formation of apatites as the material comes

into contact with phosphate-containing fluids (Sarkar

et al. 2005), via spontaneous transformation from

initial amorphous calcium phosphate phases (Tay et al.

2007, Tay & Pashley 2008).

Whereas the retentive potential of geosynthetics

(Marques 2005), concrete reinforcements (Lee et al.

2002a,b) and rigid postsystems within canal spaces

(Mitchell et al. 1994, Teixeira et al. 2006) may be

evaluated en masse using conventional pull-out test

designs, thermoplastic root filling materials and sealers

are not amendable to gripping that is a prerequisite for

this type of mechanical testing (Goracci et al. 2007).

Thus, the thin-slice push-out test has been used quite

frequently for evaluating the dislocation resistance of

root filling materials (Gesi et al. 2005, Sousa-Neto et al.

2005, Gancedo-Caravia & Garcia-Barbero 2006, Skid-

more et al. 2006, Ungor et al. 2006, Bouillaguet et al.

2007, Fisher et al. 2007, Jainaen et al. 2007, Nagas

et al. 2007, Sly et al. 2007, Ureyen Kaya et al. 2008).

The strength of that experimental design is that each

horizontal root slab being tested is derived from a root

filled canal and contains a cross-section of the thermo-

plastic root filling material and sealer to be investigated.

In the present study, a modified thin-slice push-out test

was designed to evaluate the dislocation resistance of

root canal sealers that were applied in bulk to simu-

lated canal spaces without the use of thermoplastic

material cores. The null hypothesis tested was that

there are no differences in the dislocation resistance of

three root canal sealers from radicular dentine when

the set sealers are tested with and without immersion

in a simulated body fluid (SBF).


Preparation of simulated canal spaces

Sixty intact, caries-free human canine teeth were

collected after the patients’ informed consents were

obtained under a protocol reviewed and approved by

the Human Assurance Committee of the Medical

College of Georgia, Georgia, USA. For each tooth, a

0.90 ± 0.05 mm thick longitudinal slab was prepared

by making buccolingual sections parallel to the longi-

tudinal axis of the tooth using a slow-speed diamond

saw (Isomet; Buehler Ltd, Lake Bluff, IL, USA) under

water-cooling. A Plexiglas platform containing a cylin-

drical well was affixed to the base of a mini drill press

to generate vertically oriented, truncated cavities of

uniform dimensions within the tooth slab (Fig. 1a).

A 0.6 mm drill bit was first used to prepare pilot holes

in the radicular dentine adjacent to the dental pulp.

Each pilot hole was carefully drilled so that it was

equidistant from the cementum and the canal wall.

Two pilot holes each were prepared in the coronal,

middle and apical thirds of the root.

Each hole was subsequently enlarged using a size 40,

25 mm long 0.04 taper ProFile nickel titanium rotary

instrument (Dentsply Tulsa Dental Specialties). To

ensure optimal cutting efficacy, a new instrument

Huffman et al. Dislocation resistance of ProRoot Endo Sealer


was used for each tooth slab. The drill press and the

thickness of the Plexiglas platform were configured so

that the rotary instrument penetrated the cylindrical

well to the same depth every time (Fig. 1b). This

permitted preparation of all truncated cavities to the

D16 diameter of the rotary instrument (i.e. 1.04 mm)

along the surface of the tooth slab. Inadvertent

preparation of cavities with nonvertical extrusion paths

was prevented by aligning the rotary instrument

perpendicular to the tooth slab (Fig. 1c). The experi-

mental design ensured that all cavities created in the

coronal, middle and apical thirds of the roots had

comparable dimensions. The artificial canal spaces

were also completely devoid of calcospherites that are

found along the mineralization front of the noninstr-

umented portions of natural root canal spaces. This

eliminated the issue of unpredictable augmentation in

sealer dislocation resistance that is caused by the

presence of undercuts and increased contact areas in

calcospherite-containing canal walls. The tooth slabs

were divided randomly into six groups of 10 slabs each

for evaluation of three endodontic sealers with or

without immersion in a SBF. Six cavities were created

in each tooth slab, with the two apical cavities residing

in transparent, sclerotic radicular dentine (Fig. 1d). For

each group, 20 simulated canal spaces were available

Figure 1 Experimental setup for the preparation of perpendicular truncated cavities of uniform dimensions in different locations

of a longitudinal tooth slab. (a) A mini drill press (D) with a 25 mm thick Plexiglas platform (B) affixed to its base (pointer).

(b) A tooth slab was placed over a supporting well in the Plexiglas platform. A 0.04 taper size 40 Profile nickel titanium rotary

instrument was inserted through a pre-drilled pilot hole in the tooth slab to create a truncated hole with the basal diameter

corresponding to the D16 diameter (i.e. 1.04 mm) of the rotary instrument. The drill press was set to drill exactly to the same depth

every time to ensure that each hole has the same circumference. (c) As slanted preparations are not amendable to push-out

testing, the current setup ensured that all cavities were created perpendicular to the tooth slab. (d) Two tapered cavities each were

prepared in the apical (Ap), middle (Mi) and coronal (Co) thirds of the root dentine. Pointer: cementoenamel junction; open

arrowhead: cementum.

Dislocation resistance of ProRoot Endo Sealer Huffman et al.


from each of the three respective radicular dentine

locations (n = 20).

Filling of root canal sealers

The tooth slabs were immersed in 17% ethylenedia-

mine tetra-acetic acid (EDTA) and ultrasonicated for

5 min to dissolve the smear layer created during the

hole-shaping procedures. The slabs were further

immersed in 6.15% sodium hypochlorite and ultraso-

nicated for 5 min to remove organic debris and the

demineralized collagen matrix created during EDTA

application. The rationale for en masse cleaning was to

further ensure that differences in dislocation resistance

of the sealers from different dentine locations were not

caused by inadequate cleaning of the apical radicular

dentine.

The three sealers investigated in this study were Pulp

Canal Sealer (SybronEndo; Sybron Dental Specialties

Inc., Orange, CA, USA), AH Plus Jet (Dentsply Caulk,

Milford, DE, USA) and the experimental ProRoot Endo

Sealer. The former two sealers were mixed according to

the manufacturers’ instructions. The calcium silicate-

based sealer was mixed with a liquid-to-powder ratio of

1 : 2 and covered with moist gauze to avoid evapora-

tion of the water component. All cavities from one

tooth slab were filled with one type of sealer. Each tooth

slab was placed over a Mylar strip (Angst & Pfister,

Geneva, Switzerland), which in turn was placed over a

microscope glass slide. For Pulp Canal Sealer and

ProRoot Endo Sealer, the sealer was mixed and placed

inside a 19-gauge AccuDose Needle Tube (Centrix,

Shelton, CT, USA). The sealer was dispensed into the

cavities so that each hole was filled with excess sealer.

For AH Plus Jet, the sealer was dispensed directly from

the double-barrel mixing syringe via an intraoral tip

attached to an auto-mixing tip. The surface of the tooth

slab was then covered with another Mylar strip and a

glass slide. The assembly was secured with binder clips

so that excess sealer was expressed laterally from the

surface and bottom Mylar strips. The assemblies were

transferred to humidors and stored under 100%

relative humidity for 1 week until all the sealers had

completely set.

The binder clips were released and the Mylar strips

were removed from the tooth slab to expose the set

sealers. The top and bottom surfaces of each tooth slab

were polished with 800-grit silicon carbide papers

under running water to remove the excess sealer. For

each sealer, one subgroup of 10 tooth slabs was tested

immediately after polishing, whilst the other subgroup

of 10 tooth slabs was immersed for 4 weeks at 37 �C in

a phosphate-containing SBF prior to testing. The SBF

contained 136.8 mmol L)1 NaCl, 3.0 mmol L)1 KCl,

2.5 mmol L)1 CaCl2Æ6 H2O, 1.5 mmol L)1 MgCl2Æ6-H2O, 0.5 mmol L)1 Na2SO4Æ10 H2O, 4.2 mmol L)1

NaHCO3 and 1.0 mmol L)1 K2HPO4Æ3H2O in deionized

water (pH 7.4). To prevent bacterial growth, 0.02%

sodium azide was also included in the SBF.

Dislocation resistance evaluation

The dislocation resistance of the set root canal sealers

was evaluated using a thin-slice push-out test design

(Chandra & Ananth 1995, Chandra & Ghonem 2001).

Prior to testing, the thickness of each tooth slab was

measured to the nearest 0.01 mm using a pair of

digital calipers. A 0.7 mm diameter carbon steel

cylindrical plunger was used for the push-out test.

The plunger had a clearance of about 0.1 mm from

either side of the dentinal wall when it is perfectly

aligned with the apical part of the truncated hole. The

plunger was attached to a 10 kg load cell that was

connected to a universal testing machine (Vitrodyne

V1000 universal tester; Liveco Inc., Burlington, VT,

USA). The push-out device consisted of a clear

Plexiglas platform with a vertical cylindrical channel,

which served as the support for the tooth slab and

provided space for the vertical movement of the

plunger through the truncated hole (Fig. 2a). To

ensure optimal alignment of the plunger with the

sealer-filled hole, a horizontal channel was drilled

through the Plexiglas platform into the vertical chan-

nel (Fig. 2b). A fibre-optic light guide was inserted into

the horizontal channel to provide high intensity

illumination of the truncated hole during the align-

ment procedure. Each root slab was secured with

sticky wax in an apical-coronal direction to the

supporting Plexiglas platform, so that the smaller

diameter apical side of the sealer-filled hole was facing

the plunger. Each sealer-filled hole was subjected to

compressive loading at a cross-head speed of

10 lm s)1 in order to displace the set sealer toward

the coronal aspect of the hole. As the plunger contacts

the set sealer on loading, shear stresses were intro-

duced along the sealer-dentine interface, causing the

set sealer to be dislocated from the walls of the

radicular dentine. Failure was confirmed by the

appearance of a sharp drop along the load/displace-

ment curve recorded by the testing machine. After

performing push-out testing of the first hole, the tooth

slab was carefully removed and realigned with the



second hole. The procedures were repeated until the

set sealers were dislodged from all the six cavities

within a tooth slab. After the push-out test, each root

slab was examined with a stereomicroscope at 30·magnification to determine the mode of failure. Failure

modes were classified as: adhesive failure along the

sealer-dentine interface; cohesive failure within the

sealer, and mixed failure that consisted of partial

adhesive failure along the dentinal walls and partial

cohesive failure within the sealer (Fig. 2c).

Digitized photographs of each tested hole were taken

from the coronal and apical aspects of the tooth slab

together with a millimetre scale for calibration purpose.

Such a procedure was performed after completion of

the push-out test as this permitted better contrast of the

circumference of the cavities. The circumferences of the

coronal (C) and apical aspects (A) of each cavity were

measured from the digitized images using image

analysis software (Image 4.01; Scion Corp., Frederick,

MA, USA). The area of the sealer-dentine interface was

approximated by 0.5 · (C + A) · h, where h represents

the thickness of the tooth slab. Dislocation resistance of

the sealer, as represented by the push-out strength, was

computed by dividing the maximum load (N) derived

from the load displacement curve with the sealer-

dentine interfacial area (mm2) and expressed in mega-

Pascals (MPa). The same procedures were applied to

those tooth slabs that had been immersed in SBF for

4 weeks.

Statistical analysis

Each sealer-filled hole was treated as a statistical unit.

For each of the six subgroups, data (n = 20) from the

three radicular dentine locations (i.e. coronal, middle

and apical thirds) were analysed using one-way anova

to determine if dislocation resistance of a particular

sealer was affected by the location of the sealer. As there

were no differences in the dislocation resistance amongst

dentine locations in all the six subgroups, data from the

coronal, middle and apical aspects of each subgroup

were pooled together for further analysis (n = 60). As

the pooled data were not normally distributed, log10-

transformation of the data was performed to normalize

the data before statistical evaluation. The transformed

pooled data were evaluated using a two-way anova

design, with sealer type and SBF storage as independent

variables. Post hoc pair-wise comparisons were per-

formed using Tukey multiple comparisons. The Student

paired t-test was conducted within the same sealer type

to examine if there was difference between the subgroup

that was tested without SBF immersion and the other

Figure 2 Experimental setup for the thin-slice push-out test. (a) The plunger (P) was connected to a 10 kg load cell (L). The

plunger was aligned with the cylindrical well (arrow) of a clear Plexiglas stage. The latter had a side channel (open arrowhead) for

the fitting of a fibre-optic light guide. (b) Each tooth slice was placed on top of the cylindrical well. The plunger had a diameter of

0.7 mm whilst the truncated hole had diameters of about 0.94 and 1.04 mm along its top and base. The use of light illumination

ensured that the plunger was aligned with the centre of the hole so that the sealer was pushed out without the plunger contacting

the wall of the hole. (c) Examples of adhesive failure, mixed failure and cohesive failure of the sealers, as observed through a

stereomicroscope after the push-out test.



that was tested after SBF immersion. Statistical signifi-

cance was set at a = 0.05.

Scanning electron microscopy

After push-out testing, two slabs from each of the six

subgroups were air-dried, sputter-coated with gold/

palladium, and examined using a field emission scan-

ning electron microscope (Model XL-30 FEG; Philips,

Eindhoven, The Netherlands) at 15 KeV. The objective

of the morphologic examination was not to reiterate

the assessment of failure modes that had been per-

formed using stereomicroscopy. Rather, the higher

resolution of a field emission microscope was utilized to

substantiate whether calcium phosphate-like phases

and their phase transformation could be identified after

the calcium silicate-based sealer was immersed in the

phosphate-containing SBF.

Results

Representative load–displacement curves of the three

root canal sealers are shown in Fig. 3a. Despite the

differences in the magnitude of the maximum load

achieved in the three sealers, their load–displacement

curves demonstrated were characterized by four

regions. There was an initial linear increase in load

(zone I) that corresponded with the increase in shear

stresses along the sealer-dentine interfaces as

the compressive load was applied from the base of the

inverted truncated sealer core. Prior to reaching the

maximum compressive load, the shear stresses reached

a critical value whereupon delamination was initiated

from the top of the inverted core. The increase in

Poisson’s ratio along the nondelaminated part of the

core (i.e. expansion) resulted in increased work to

continue the delamination and hence a change in the

slope of the load–displacement curve (zone II). Upon

reaching the maximum load, propagation of shear

stresses toward the bottom of the interface resulted in

complete interfacial delamination and a sudden sharp

drop in the recorded load (zone III). During the final

push-out phase (zone IV), resistance to displacement by

sliding friction and surface roughness of the delami-

nated sealer core resulted in a progressive, less abrupt

decline in the recorded load as the delaminated core

was displaced out of the truncated hole.

For each sealer with or without SBF immersion, no

significant differences were observed amongst the push-

out strengths obtained from different dentine locations

(Fig. 3b). Thus, data from the apical, middle and

coronal thirds of the roots were pooled to provide a

more robust analysis of the effects of sealer type and

SBF immersion on push-out strengths (Fig. 3c). When

the specimens were tested without SBF immersion,

significant differences (P < 0.001) were observed

amongst the three sealers, with the calcium silicate-

based sealer producing the highest push-out strength

(16.2 ± 6.5 MPa) followed by AH Plus Jet (3.5 ±

1.7 MPa) and Pulp Canal Sealer (0.7 ± 0.6 MPa) in

decreasing order. Significant differences in push-out

strength was also observed for specimens that were

tested after they were immersed in SBF for 4 weeks

(P < 0.001), following the same order as previously

described (calcium silicate-based sealer 22.4 ±5.0 MPa;

AH Plus Jet 6.6 ± 1.7 MPa; Pulp Canal Sealer

0.4 ± 0.3 MPa). Interaction of these two factors were

also significant (P < 0.001). For the AH Plus Jet and

the calcium silicate-base sealer, Student paired t-tests

revealed significant differences (P < 0.05) between the

push-out strengths generated from specimens that were

tested without SBF immersion and those that were

tested after immersion in SBF.

The per cent distribution of failure modes amongst

the six subgroups is presented in Fig. 4. No cohesive

failure was observed for Pulp Canal Sealer. This sealer

also exhibited an increase in the percentage of adhesive

failure after storage in SBF. A preponderance of mixed

failures was seen in AH Plus under the two storage

conditions, whilst cohesive failures within the sealer

were predominantly identified for the calcium silicate-

based sealer.

Under scanning electron microscopy, failures classi-

fied as adhesive failures in the Pulp Canal Sealer groups

invariably contained some sealer remnants along the

dentinal walls (not shown). However, the overall

impressions of those dentinal walls were still relatively

smooth when compared with the mixed failures

observed in the other sealer groups. A cohesive failure

in AH Plus Jet after SBF immersion is shown in Fig. 5a.

A high magnification view of the fractured sealer

surface revealed characteristic multi-faceted fillers that

were partially embedded, amongst other smaller fillers,

within a resinous matrix (Fig. 5b). A mixed failure

mode in the calcium silicate-based sealer after SBF

immersion is depicted in Fig. 6a. Spherical bodies were

identified along the sealer–dentin interface as well as

the surface of the fractured sealer (Fig. 6b). These

spherical phases were not observed from fractured

specimens of the same sealer that had not been

immersed in SBF (not shown). Very high magnification

views of the specimens that had been immersed in SBF



before testing revealed phase transformation of the

spherical bodies to spherules with clustered polycrys-

talline surfaces (Fig. 6c). Individual crystallites that

protruded from the surface of these spherules were

about 40–70 nm in diameter (Fig. 6d).

Discussion

This study utilized a modified push-out protocol that

was designed specifically to examine the retentive

potential of sealers in radicular dentine. Although the

study design is far removed from clinical practice, the

results indicate that under identical cleaning and

shaping conditions that may not be easily achieved

under a clinical setting, the dislocation resistance of a

particular sealer is independent of the location of the

radicular dentine. Moreover, the dislocation resistance

of the three sealers were significantly different from

each other and that two of the three sealers exhibited

higher dislocation resistance after immersion in SBF.

Thus, the null hypothesis has to be rejected. Although

a modified push-out test design was used in this study,

it is interesting to note that the relatively low push-out

strengths for AH Plus and Pulp Canal Sealer were

similar to the range reported for similar sealers

(2.00 ± 0.65 MPa for AH Plus and 0.79 ± 0.52 MPa

for Kerr EWT sealer) in a previous study (Fisher et al.

2007).

Although testing designs that involve the use of

natural canal spaces have obvious pragmatic appeal to

clinicians, there are severe limitations from a materials

science perspective. First, the application of a compres-

sive load on top of a thermoplastic material, which has

Load-displacement curves of endodontic sealers(a)

(b)

(c)

Push out strength of endodontic sealers

Push out strength of endodontic sealers(pooled results; N = 60)

Displacement (μm)

Ap: Apical thirdMi: Middle thirdCo: Coronal third

Before immersion in SBFAfter immersion in SBF for4 weeks

Before immersion in SBFAfter immersion in SBF for4 weeks

70

60

50

40 I

II

III

IV30

20

10

00 200 400

ProRoot Endo SealerAH Plus Jet Pulp Canal Sealer

600 800 1000

30

25

20

15

10

5

0

Lo

ad (

N)

MP

a

25

20

15

10

5

0Pulp Cancal

SealerAH Plus Jet ProRoot Endo

Sealer

Pulp Cancal Sealer

Ap Mi Co Ap Mi Co Ap Mi CoAH Plus Jet ProRoot Endo

Sealer

3 C

2

B

1A

Dis

loca

tio

n r

esis

tan

ce (

MP

a)

Figure 3 Push-out strength results. (a) Representative load-

displacement curves of the three sealers that were tested in

bulk without an accompanying gutta-percha core. Load is

expressed as Newtons (N) and displacement is expressed as

microns (lm). Zone I: initial linear increase in load; zone II:

change in slope of the load–displacement curve before

reaching maximum load; zone III: initial sudden sharp drop

in recorded load upon interfacial delamination; zone IV: final

push-out phase. When magnified, the four regions described

for ProRoot Endo Sealer could also be seen in the load–

displacement curves of AH Plus Jet and Pulp Canal Sealer. (b)

dislocation resistance (expressed as MPa) of the three sealers in

the apical third (Ap), middle third (Mi) and coronal third (Co)

of the root dentine with and without storage in a simulated

body fluid (SBF) (n = 20/location/storage subgroup). As there

were no statistical differences in the push-out strengths of each

sealer amongst different locations at each time period, data

from the three locations were pooled (n = 60) for subsequent

statistical comparisons. (c) The pooled data was analysed

using a two-way anova design with sealer type and SBF

storage as independent variables. For specimens tested without

SBF immersion, sealers with different numerals above their

corresponding data columns represent significant differences

(P < 0.001). For specimens that were tested after they were

immersed in SBF, sealers with different upper case letters

above their corresponding data columns represent significant

differences (P < 0.001). For each sealer type, a horizontal bar

above the respective columns for the two immersion protocols

indicates no statistical difference (P > 0.05).



the tendency to flow during testing generates results

that are susceptible to erroneous interpretation. Unless

the rheological properties of the materials being com-

pressed are equivalent (Kohyama et al. 2003, Tornqvist

et al. 2004), statistical comparison of the results

derived from two thermoplastic root filling materials

is virtually meaningless. This could also have been

responsible, in part, for the recent report that sealers

tested in thin films using the thin-slice push-out test

were considerably weaker than when the same sealer

Failure modes of push out tests

Before immersionin SBF

PulpCanalSealer

AH PlusJet

ProRootEndoSealer

61.7%38.3%

20%

80%

6.7%

36.6%

56.7%74.1%

8.6%17.3%

100% 95%

5%

After immersion in SBFfor 4 weeks

Adhesive Mixed Cohesive

Figure 4 Distribution of adhesive, mixed

and cohesive failures of the three sealers

in specimens that were tested without

immersion in simulated body fluid (SBF)

and specimens that were tested after

immersion in SBF for 4 weeks.

Figure 5 Scanning electron microscopy (SEM) of AH Plus after immersion in simulated body fluid (SBF) and push-out testing. (a)

Low magnification SEM of a cohesive failure mode exhibited by a specimen from the AH Plus Jet group after SBF immersion. (b) A

higher magnification view showing the presence of large, multi-facet fillers (open arrowheads) that are characteristic of the AH

Plus sealer. These fillers were embedded in a resinous matrix together with other fine filler particles.



was tested in bulk by eliminating the thermoplastic

core material from the canal space (Jainaen et al.

2007). The important results generated by those

authors provided incontestable substantiation that the

so-called ‘push-out bond strength’ produced by the

conventional thin-slice push-out test is not a material

property. It is prudent to emphasize that the mechan-

ical and physical properties of engineering and bioma-

terials such as flexural strength, fracture toughness or

melting point should exhibit a consistent range of

values under identical testing conditions (Callister

1994).

To minimize the shortcoming of applying a com-

pression stress over a compliant material, the largest

plunger that corresponds to the size of the thermoplas-

tic root filling material is usually selected for the thin-

slice push-out test (Gesi et al. 2005, Bouillaguet et al.

2007) Whilst this is a legitimate compromise, the

procedure succinctly requires the use of different

diameter plungers for different depths of a tapered root

canal. As the contact surface areas of the plungers are

different, data generated from different parts of the

canal walls are nonstandardized. Thus, it is futile to

statistically compare the results generated by a con-

ventional thin-slice push-out test from the coronal

third, versus those generated from the middle and

apical thirds of the canal walls.

The third limitation involves the testing of root

fillings that comprise multiple, nonuniform interfaces.

Whilst the uneven distribution of stress fields around

Figure 6 Scanning electron microscopy (SEM) of ProRoot Endo Sealer after immersion in simulated body fluid (SBF) and push-out

testing. (a) Low magnification SEM of a mixed failure mode exhibited by a specimen from the ProRoot Endo Sealer group after SBF

immersion. Sealer remnants (pointer) could be seen on part of the wall. The remaining part of the walls was devoid of sealer

remnants and appeared comparatively smooth (arrow). (b) A higher magnification view of the sealer remnants, showing the

presence of spherical bodies on sealer surface. These spherical bodies were previously shown to be amorphous calcium phosphate-

like spheres that were formed by the reaction of calcium hydroxide released by the calcium silicate with the phosphate ions present

in the SBF. (c) A very high magnification view showing that some of the amorphous calcium phosphate-like spheres (arrow) were

spontaneously transformed into apatite-like clusters along the surface of those spheres (pointer). (d) A close-up view of a spherical

apatite-like cluster showing the presence of individual apatite-like crystallites (open arrowheads).



interfaces with variable circumferential thickness can-

not be over-emphasized (Shirazi-Adl & Forcione 1992,

Mequid & Zhu 1995), the uncertainty with respect to

which interface was consistently dislodged imposes

rigorous challenges when specific hypotheses such as

the dislocation resistance of sealers from radicular

dentine are to be tested.

The fourth limitation is that one is almost certain to

find noninstrumented areas that co-exist with instru-

mented areas in an oval-shaped canal that has been

cleaned and shaped (Peters 2004). For the noninstru-

mented areas that are treated with sodium hypochlorite

as an irrigant, one should expect increases in both

undercut retention and surface contact areas within

the calcospherite-containing regions (Wakabayashi

et al. 1993, Tatsuta et al. 1999) that inadvertently

augments the dislocation resistance of the sealer being

investigated. For example, comparing the results gen-

erated from a natural canal space with 50% noninstr-

umented canal walls versus one that has 20%

noninstrumented canal walls may result in erroneous

conclusions on the dislocation resistance of various

sealers from radicular dentine. It is unrealistic to

quantify the extent of noninstrumented natural canal

walls from a root slab either before or after a push-out

test. Because of these limitations, a modified push-out

strength testing design was utilized in the present

study.

Even without SBF immersion, the calcium silicate-

based sealer was approximately 16 times as difficult to

be dislodged from the radicular dentine walls as Pulp

Canal Sealer, and almost four times as resistant to

dislodging as AH Plus Jet. This may be due, in part, to

the hardness of the calcium silicate-based sealer after

setting in the presence of 100% relative humidity. As

natural root canals cannot be completely dehydrated

(Amyra et al. 2000, Hosoya et al. 2000) due to the

retention of moisture within the dentinal tubules,

similar hardness should be expected of the set sealer

when it is used for filling natural canals. The tenacity

of this sealer to radicular dentine cannot be solely

attributed to sealer penetration into the dentinal

tubules following depletion of the smear layer, as the

dentine from the apical third of the roots is often highly

sclerotic. It is beyond the scope of this study to provide

definitive annotations on whether the increased dislo-

cation resistance is caused by the frictional resistance

or micromechanical/chemical adhesion of the sealer to

dentine (Shirazi-Adl 1992, Goracci et al. 2005). This

issue should be further investigated in the future using

more advanced transmission electron microscopy and

chemoanalytical techniques. Nevertheless, the

increased dislocation resistance of the calcium

silicate-based sealer to radicular dentine should be

advantageous in maintaining the integrity of the

sealer–dentine interface during tooth flexure, as well

as during the preparation of postholes within the filled

canal spaces.

The concern on whether the dislocation resistance of

root canal sealers is adversely affected by the contam-

ination of body fluids was simulated in the present

study by immersing the specimens in a SBF. This is an

exaggerated simulation as the entire tooth slab was

immersed in the SBF after the cavities were filled with

sealers. The increase in dislocation resistance of the AH

Plus Jet is probably caused by swelling of the epoxy

resin component after water sorption (Fernandez-Gar-

cıa & Chiang 2002, Domotor & Hentschke 2004). For

the calcium silicate-based sealer, continuous matura-

tion of the material (Andriamanantsilavo & Amziane

2004) may also have increased the material’s disloca-

tion resistance. However, the occurrence of spherical

phases along the sealer–dentine interface and within

the remnant fractured sealer after the specimens were

immersed in the phosphate-containing SBF is notable.

These spherical phases have previously been identified

as amorphous calcium phosphate when Portland

cement was immersed in a phosphate-containing fluid

(Tay et al. 2007). Amorphous calcium phosphate

phases undergo spontaneous transformation to car-

bonated apatites (Gadaleta et al. 1996), producing

hollow spherules of apatite clusters (Eanes 2001, Tay

& Pashley 2008) that contributed to the ex vivo

bioactivity of calcium silicate-containing materials

when they interact with phosphate ions. Similar

apatite-containing clusters had been observed when

Mineral Trioxide Aggregate was immersed in phos-

phate-containing fluids (Sarkar et al. 2005). The apat-

itic composition in these spherules has also been

established using x-ray diffraction (XRD) and Fourier

transform-infrared spectroscopy (FT-IR) (Tay et al.

2007). No attempt was made to analyse the compar-

atively smooth spherical phases and the crystallite-

containing spherules in this study, as these phases were

present adjacent to calcium-phosphate rich dentine and

on the surface of the fractured sealer. The use of energy

dispersive X-ray analysis to analyse these surface

phases would have yielded information that includes

the subsurface elemental composition of the dentine

and sealer components. Likewise, these phases were not

amendable for collection and purification for XRD and

FT-IR analyses. Thus, they are only referred to as



amorphous calcium phosphate-like and apatite-like in

the present study. Generation of these reaction phases

only in specimens that were immersed in the SBF could

also have resulted in the increase in frictional resis-

tance of the sealer–dentin interface. Although it is

presumptuous to correlate the ‘in vitro bioactivity’ (i.e.

the ability to form carbonate hydroxyapatite on the

surface of a biomaterial when it is exposed to SBF)

(LeGeros 2002, Zhao et al. 2005, Panzavolta et al.

2008) observed in the present study with ‘clinical

bioactivity’ (i.e. the property of the material to develop

a direct, adherent and strong bonding with the bone

tissue) (Hench et al. 1978, Hench 1994), the issue of

‘clinical bioactivity’ associated with the use of endo-

dontic sealers in general is of practical clinical interest

and should be duly investigated.

Conclusion

Within the limits of the modified push-out testing

design utilized in the present ex vivo study, it may be

concluded that:

• Under identical cleaning and shaping conditions,

the dislocation resistance of ProRoot Endo Sealer, AH

Plus Jet and Pulp Canal Sealer are independent of the

location of the radicular dentine.

• The dislocation resistance of the three sealers are in

descending order: ProRoot Endo Sealer, AH Plus Jet and

Pulp Canal Sealer.

• Both ProRoot Endo Sealer and AH Plus Jet exhibited

higher dislocation resistance after immersion in a SBF.

• ProRoot Endo Sealer exhibited amorphous calcium

phosphate-like phases that spontaneously transformed

into apatite-like phases after immersion in the phos-

phate-containing SBF. This phenomenon probably

accounts for the in vitro bioactivity of this calcium

silicate-based sealer.

Acknowledgements

This study was supported by Dentsply Tulsa Dental

Specialties. Dr Primus and Dr Gutmann served as

consultants for Dentsply Tulsa Dental Specialties. The

authors are grateful to Miss Anna Lam for her

secretarial support.

References

Amyra T, Walsh LT, Walsh LJ (2000) An assessment of

techniques for dehydrating root canals using infrared laser

radiation. Australian Endodontic Journal 26, 78–80.

Andriamanantsilavo NR, Amziane S (2004) Maturation of

fresh cement paste within 1- to 10-m-large formworks.

Cement and Concrete Research 34, 2141–52.

Bouillaguet S, Bertossa B, Krejci I, Wataha JC, Tay FR, Pashley

DH (2007) Alternative adhesive strategies to optimize

bonding to radicular dentin. Journal of Endodontics 33,

1227–30.

Callister WD Jr. (1994) Materials Science and Engineering: an

Introduction, 3rd edn. New York: JohnWiley & Sons, pp. 1–5.

Camilleri J, Pitt Ford TR (2006) Mineral trioxide aggregate: a

review of the constituents and biological properties of the

material. International Endodontic Journal 39, 747–54.

Chandra N, Ananth CR (1995) Analysis of interfacial behavior

in MMCs and IMCs using thin slice push-out tests. Compos-

ites Science and Technology 54, 87–100.

Chandra N, Ghonem H (2001) Interfacial mechanics of push-

out tests: theory and experiments. Composites Part A: Applied

Science and Manufacturing 32, 578–84.

De Moor R, Hommez G (2000) The importance of apical and

coronal leakage in the success or failure of endodontic

treatment. Revue Belge de Medecine Dentaire 55, 334–44.

Domotor G, Hentschke R (2004) Equilibrium swelling of an

epoxy-resin in contact with water – a molecular dynamics

simulation study. Macromolecular Theory and Simulations 13,

506–11.

Eanes ED (2001) Amorphous calcium phosphate. In: Chow LC,

Eanes ED, eds. Monographs in Oral Science, Vol. 18, Basel:

Karger, pp. 130–47.

Fernandez-Garcıa M, Chiang MYM (2002) Effect of

hygrothermal aging history on sorption process, swelling,

and glass transition temperature in a particle-filled

epoxy-based adhesive. Journal of Applied Polymer Science

84, 1581–91.

Fisher MA, Berzins DW, Bahcall JK (2007) An in vitro

comparison of bond strength of various obturation materials

to root canal dentin using a push-out test design. Journal of


Gadaleta SJ, Paschalis EP, Betts F, Mendelsohn R, Boskey AL

(1996) Fourier transform infrared spectroscopy of the

solution-mediated conversion of amorphous calcium phos-

phate to hydroxyapatite: new correlations between X-ray

diffraction and infrared data. Calcified Tissue International 58,

9–16.

Gancedo-Caravia L, Garcia-Barbero E (2006) Influence of

humidity and setting time on the push-out strength of

mineral trioxide aggregate obturations. Journal of Endodon-

tics 32, 894–6.

Gesi A, Raffaelli O, Goracci C, Pashley DH, Tay FR, Ferrari M

(2005) Interfacial strength of Resilon and gutta-percha to

intraradicular dentin. Journal of Endodontics 31, 809–13.

Gettleman BH, Messer HH, ElDeeb ME (1991) Adhesion of

sealer cements to dentin with and without the smear layer.


Goracci C, Fabianelli A, Sadek FT, Papacchini F, Tay FR,

Ferrari M (2005) The contribution of friction to the



dislocation resistance of bonded fiber posts. Journal of


Goracci C, Grandini S, Bossu M, Bertelli E, Ferrari M (2007)

Laboratory assessment of the retentive potential of adhesive

posts: a review. Journal of Dentistry 35, 827–35.

Gou Z, Chang J, Zhai W, Wang J (2005) Study on the self-

setting property and the in vitro bioactivity of beta-Ca2SiO4.

Journal of Biomedical Materials Research. Part B, Applied

Biomaterials 73, 244–51.

He RX, Yan SG, Wu LD, Wang XH, Dai XS (2007) Position of

the prosthesis and the incidence of dislocation following

total hip replacement. Chinese Medical Journal (English) 120,

1140–4.

Hench LL (1994) Bioceramics: from concept to clinic. Journal

of the American Ceramics Society 74, 1487–510.

Hench LL, Splinter RJ, Allen WC, Greenlee TK (1978) Bonding

mechanisms at the interface of ceramic prosthetic materials.

Journal of Biomedical Materials Research 2, 117–41.

Hosoya N, Nomura M, Yoshikubo A, Arai T, Nakamura J, Cox

CF (2000) Effect of canal drying methods on the apical seal.


Jainaen A, Palamara JE, Messer HH (2007) Push-out bond

strengths of the dentine-sealer interface with and without a

main cone. International Endodontic Journal 40, 882–90.

Kohyama K, Sasaki T, Dan H (2003) Active stress during

compression testing of various foods measured using a

multiple-point sheet sensor. Bioscience, Biotechnology, and

Biochemistry 67, 1492–8.

Laghios CD, Cutler CW, Gutmann JL (2000) In vitro evidence

that lipopolysaccharide of an oral pathogen leaks from root-

end filled teeth. International Endodontic Journal 33, 333–9.

Lee HS, Noguchi T, Tomosawa F (2002a) Evaluation of the

bond properties between concrete and reinforcement as a

function of the degree of reinforcement corrosion. Cement

and Concrete Research 32, 1313–8.

Lee KW, Williams MC, Camps JJ, Pashley DH (2002b)

Adhesion of endodontic sealers to dentin and gutta-percha.


LeGeros RZ (2002) Properties of osteoconductive biomaterials:

calcium phosphates. Clinical Orthopaedics and Related

Research 395, 81–98.

Madison S, Wilcox LR (1988) An evaluation of coronal

microleakage in endodontically treated teeth. Part III. In

vivo study. Journal of Endodontics 14, 455–8.

Marques JMM (2005) Finite element modelling of the pull-out

test of geosynthetics. In: Onate E, Owen DRJ, eds. VIII

International Conference on Computational Plasticity COMPLAS

VIII. Barcelona: CIMNE, pp. 1–4.

Mequid SA, Zhu ZH (1995) Stress distribution in dissimilar

materials containing inhomogeneities near the interface

using a novel finite element. Finite Elements in Analysis and

Design 20, 283–98.

Mitchell CA, Orr JF, Connor KN, Magill JP, Maguire GR (1994)

Comparative study of four glass ionomer luting cements

during post pull-out tests. Dental Materials 10, 88–91.

Munoz HR, Saravia-Lemus GA, Florian WE, Lainfiesta JF

(2007) Microbial leakage of Enterococcus faecalis after post

space preparation in teeth filled in vivo with RealSeal versus

gutta-percha. Journal of Endodontics 33, 673–5.

Nagas E, Cehreli ZC, Durmaz V, Vallittu PK, Lassila LV (2007)

Regional push-out bond strength and coronal microleakage

of Resilon after different light-curing methods. Journal of


Panitvisai P, Messer HH (1995) Cuspal deflection in molars in

relation to endodontic and restorative procedures. Journal of


Panzavolta S, Torricelli P, Sturba L, Bracci B, Giardino R,

Bigi A (2008) Setting properties and in vitro bioactivity

of strontium-enriched gelatin–calcium phosphate bone

cements. Journal of Biomedical Materials Research A 84,

965–72.

Peters OA (2004) Current challenges and concepts in the

preparation of root canal systems: a review. Journal of


Saleh IM, Ruyter IE, Haapasalo MP, Ørstavik D (2003)

Adhesion of endodontic sealers: scanning electron micros-

copy and energy dispersive spectroscopy. Journal of End-

odontics 29, 595–601.

Salehrabi R, Rotstein I (2004) Endodontic treatment outcomes

in a large patient population in the USA: an epidemiological

study. Journal of Endodontics 30, 846–50.

Sarkar NK, Caicedo R, Ritwik P, Moiseyeva R, Kawashima I

(2005) Physicochemical basis of the biologic properties of

mineral trioxide aggregate. Journal of Endodontics 31, 97–

100.

Scifert CF, Brown TD, Lipman JD (1999) Finite element

analysis of a novel design approach to resisting total hip

dislocation. Clinical Biomechanics (Bristol, Avon) 14, 6697–

703.

Shirazi-Adl A (1992) Finite element stress analysis of a push-

out test. Part 1: fixed interface using stress compatible

elements. Journal of Biomechanical Engineering 114, 111–8.

Shirazi-Adl A, Forcione A (1992) Finite element stress analysis

of a push-out test. Part II: free interface with nonlinear

friction properties. Journal of Biomechanical Engineering 114,

155–61.

Skidmore LJ, Berzins DW, Bahcall JK (2006) An in vitro

comparison of the intraradicular dentin bond strength of

Resilon and gutta-percha. Journal of Endodontics 32, 963–6.

Sly MM, Moore BK, Platt JA, Brown CE (2007) Push-out bond

strength of a new endodontic obturation system (Resilon/

Epiphany). Journal of Endodontics 33, 160–2.

Sousa-Neto MD, Silva Coelho FI, Marchesan MA, Alfredo E,

Silva-Sousa YT (2005) Ex vivo study of the adhesion of an

epoxy-based sealer to human dentine submitted to irradia-

tion with Er:YAG and Nd:YAG lasers. International Endo-

dontic Journal 38, 866–70.

Tagger M, Tagger E, Tjan AH, Bakland LK (2003) Shearing

bond strength of endodontic sealers to gutta-percha. Journal




Tatsuta CT, Morgan LA, Baumgartner JC, Adey JD (1999)

Effect of calcium hydroxide and four irrigation regimens on

instrumented and uninstrumented canal wall topography.


Tay FR, Pashley DH (2008) Guided tissue remineralisation of

partially demineralised human dentine. Biomaterials 29,

1127–37.

Tay FR, Pashley DH, Rueggeberg FA, Loushine RJ, Weller RN

(2007) Calcium phosphate phase transformation produced

by interaction of the Portland cement component of white

MTA with a phosphate-containing fluid. Journal of Endodon-

tics 33, 1347–51.

Teixeira EC, Teixeira FB, Piasick JR, Thompson JY (2006) An

in vitro assessment of prefabricated fiber post systems.

Journal of the American Dental Association 137, 1006–12.

Tilashalski KR, Gilbert GH, Boykin MJ, Shelton BJ (2004) Root

canal treatment in a population-based adult sample: status

of teeth after endodontic treatment. Journal of Endodontics

30, 577–81.

Timpawat S, Harnirattisai C, Senawongs P (2001) Adhesion of

a glass-ionomer root-canal sealer to the root-canal wall.


Tornqvist R, Sunderland P, Manson JAE (2004) Determination

of the rheological properties of thermoplastic composites for

compression flow molding. Polymer Composites 21, 779–88.

Ungor M, Onay EO, Orucoglu H (2006) Push-out bond

strengths: the Epiphany–Resilon endodontic obturation

system compared with different pairings of Epiphany,

Resilon, AH Plus and gutta-percha. International Endodontic

Journal 39, 643–7.

Ureyen Kaya B, Kececi AD, Orhan H, Belli S (2008) Micro-

push-out bond strengths of gutta-percha versus thermo-

plastic synthetic polymer-based systems – an ex vivo study.


Wakabayashi H, Matsumoto K, Nakamura Y, Shirasuka T

(1993) Morphology of the root canal wall and arrangement

of underlying dentinal tubules. International Endodontic

Journal 26, 153–8.

Wang X, Sun H, Chang J (2008) Characterization of

Ca(3)SiO(5)/CaCl(2) composite cement for dental applica-

tion. Dental Materials 24, 74–82.

Weale AE, Feikes J, Prothero D, O’Connor JJ, Murray D,

Goodfellow J (2002) In vitro evaluation of the resistance to

dislocation of a meniscal-bearing total knee prosthesis

between 30� and 90� of knee flexion. Journal of Arthroplasty17, 475–83.

Wennberg A, Ørstavik D (1990) Adhesion of root canal sealers

to bovine dentine and gutta-percha. International Endodontic

Journal 23, 13–9.

Zhao W, Wang J, Zhai W, Wang Z, Chang J (2005) The self-

setting properties and in vitro bioactivity of tricalcium

silicate. Biomaterials 26, 6113–21.



Influence of post fit and post length on fractureresistance

L. Buttel, G. Krastl, H. Lorch, M. Naumann, N. U. Zitzmann & R. WeigerDepartment of Periodontology, Endodontology and Cariology, School of Dentistry, University of Basel, Basel, Switzerland

Abstract

Buttel L, Krastl G, Lorch H, Naumann M, Zitzmann NU,

Weiger R. Influence of post fit and post length on fracture

resistance. International Endodontic Journal, 42, 47–53, 2009.

Aim To investigate (i) the impact of post fit (form-

congruence) and (ii) the influence of post length on the

fracture resistance of severely damaged root filled

extracted teeth.

Methodology Ninety-six single-rooted human

teeth were root filled and divided into four groups

(n = 24 per group). Post spaces were prepared with a

depth of 6 mm (group 1, 3) and 3 mm (group 2, 4).

Form-congruence with a maximal fit of the post

within the root canal space was obtained in groups 1

and 2, whereas there was no form-congruence in

groups 3 and 4. In all groups, glass fibre reinforced

composite (FRC) posts were adhesively cemented and

direct composite crown build-ups were fabricated

without a ferrule. After thermo-mechanical loading

(1200000·, 5–50 �C), static load was applied until

failure. Loads-to-failure [in N] were compared

amongst the groups.

Results Post fit did not have a significant influence on

fracture resistance, irrespective of the post length. Both

groups with post insertion depths of 6 mm resulted in

significantly higher mean failure loads (group 1, 394 N;

group 3, 408 N) than the groups with post space

preparation of 3 mm (group 2, 275 N; group 4, 237 N).

Conclusions Within the limitations of this study,

the fracture resistance of teeth restored with FRC posts

and direct resin composite crowns without ferrules was

not influenced by post fit within the root canal. These

results imply that excessive post space preparation

aimed at producing an optimal circumferential post fit

is not required to improve fracture resistance of roots.

Keywords: endodontic post, form-congruence, frac-

ture resistance, in vitro study, post space.

Received 20 May 2008; accepted 22 September 2008

Introduction

As root filled teeth often have insufficient coronal tooth

structure, placement of a post is occasionally necessary

to provide adequate retention for the core and final

restoration. Alternatives to cast post-and-cores have

been developed and include the use of pre-fabricated

posts and custom-made cores with composite that

facilitate a chair-side restorative procedure (Heydecke

et al. 2002). In particular, fibre-reinforced composite

(FRC) posts luted with adhesive materials have become

more popular because of their favourable mechanical

and aesthetic properties. For example, the elastic

modulus of FRC posts is close to that of dentine, and

results in the stress transmitted by a fibre post to the

root dentine being lower than that caused by other

materials such as titanium or zirconia (Duret et al.

1990). There is a controversy as to whether stress

transmission and post rigidity has an impact on the

fracture resistance and/or failure mode of root filled

teeth with posts (Isidor et al. 1996, Akkayan & Gulmez

2002, Fokkinga et al. 2004). In addition to the

presence of a post, other factors possibly influencing

the load capability of root filled teeth are tooth

morphology, restorative techniques and crucially the

amount of tooth tissue lost (Trope et al. 1985, Gut-

mann 1992, Sornkul & Stannard 1992, Fernandes &

Dessai 2001).

Correspondence: Leonard Buttel, Department of Periodontol-

ogy, Endodontology and Cariology, School of Dentistry,

University of Basel, Hebelstrasse 3, CH-4056 Basel, Switzer-

land (Tel.: +41 61 2672623; fax: +41 61 2672659; e-mail:

[email protected]).

doi:10.1111/j.1365-2591.2008.01492.x


When placing posts in accordance with standard

clinical protocols, pilot drills are used to create a form-

congruent root canal up to the apical third of the root

to ensure primary post fit and retention. This optimized

post fit can be termed ‘form-congruence’ (Schmage

et al. 2005) and aims to create maximal adaptation of

the post to the surrounding root canal walls with a

thin and even post dentine cement interface. It is

believed that form-congruence facilitates stress distri-

bution along the canal wall during clinical function

(Morgano 1996). Schmage et al. (2005) examined the

form-congruence of five pre-fabricated titanium posts

luted with zinc phosphate cement and found that the

mean cement gap varied between 33 and 62 lm,

depending on the post system. For root filled teeth with

cast post-and-cores and crowns luted with zinc phos-

phate cement, a significant increase in fracture resis-

tance was reported when a maximum adaptation of a

tapered post to the residual root structure was present

(Sorensen & Engelman 1990). This effect was not

observed when parallel-sided posts were used. Prepar-

ing post spaces, however, poses several risks. The

individual curvature and cross-section of the root

canal may interfere with this preparation and create

additional weakening of the root or even root perfo-

ration. Lang et al. (2006) investigated the impact of

endodontic procedures on the deformation of anterior

maxillary teeth and found that their stability decreases

with every stage of the root canal preparation.

A significant decrease in stability was observed when

the post space was prepared, particularly following the

transformation of the conical post preparations to a

cylindrical form. It was concluded that if excessive

amounts of tooth structure are removed and the

natural geometry of the root canals are altered, this

will have a destabilizing effect on root filled teeth.

A recent study using computational, experimental and

fractographic analyses has substantiated the impact of

so called inner dentine located adjacent to the root

canal on fracture resistance of teeth (Kishen et al.

2004). Obviously, it is not only the thickness of the

dentine wall that stabilizes the root but also the

presence of inner dentine with a lower elastic modulus

than the more mineralized outer dentine. Particularly

in irregular root canals with an oval cross-section,

large diameter drills are required to ensure a circum-

ferential post fit, and thereby excessive amounts of

inner dentine are removed. Selecting a post that

corresponds best to the natural root canal diameter

without preparation, however, aims to preserve the

inner dentine substance and may be associated with a

loose-fitting post in irregular canals (no form-congru-

ence).

As soon as posts are luted adhesively to the root

canal walls, an ideal post fit within the canal (form-

congruence) is probably less important as any spaces

are filled with the luting composite. However, shrink-

age of the thicker resin cement film by nonfitting posts

may impair the clinical performance in the long term.

Otherwise, even after standardized post space prepara-

tion (using the post hole drills supplied by the

manufacturers) and optimal bonding procedures, the

high-cavityconfiguration factor may lead to gap

formation either along the cement-dentine interface

or the cement-post interface (Pirani et al. 2005). To

reduce the thickness of resin cement in irregular post

spaces, Grandini et al. (2003, 2005) suggested a

pre-cementation relining of the post with flowable

composite (anatomical post) for the cementation of fibre

posts to improve its fit to the canal space. In the light of

this background, the use of adhesive techniques for post

cementation and a minimal invasive post space prep-

aration minimizing the loss of hard tissue are clinically

preferable.

The aim of the present investigation was to study the

influence of the form-congruence of adhesively luted

glass FRC posts and of post length on the fracture

resistance of root filled teeth. The null hypothesis was

that (i) providing a form-congruence between post and

post space preparation and/or (ii) reducing the post

length would have no influence on the fracture load of

root filled teeth restored with adhesively luted glass FRC

posts and direct composite crowns.


Ninety-six extracted single-rooted human teeth (max-

illary lateral incisors and mandibular second premo-

lars) were selected that fulfilled the following criteria:

straight, sound roots, completely formed apices,

absence of root caries and no visible fracture lines

along the root. Teeth with similar dimensions at the

cementoenamel junction (CEJ) in terms of root diameter

and thickness of the dentine wall were distributed

equally amongst the four groups. The teeth were stored

in 0.1% thymol solution until further processing. The

clinical crowns were removed 1 mm below the buccal

CEJ using a diamond bur, leaving a root length of

13 ± 1 mm. All roots were cleaned with scalers.

Root canal preparation was performed using NiTi

rotary instruments (Race, FKG, La Chaux-de-Fonds,

Switzerland) under intermittent rinsing with 1%

Form-congruence of endodontic posts Buttel et al.


sodium hypochlorite to an apical size 45. The canals

were then dried with paper points and filled with

vertically compacted gutta-percha (Obtura II, Obtura

Corp, Fenton, MO, USA) using an epoxy sealer (AH

plus, Dentsply De Trey, Konstanz, Germany).

For each group, pre-fabricated glass FRC posts (FRC

Postec, Ivoclar Vivadent, Schaan, Liechtenstein) with a

9.3% taper were used. Post spaces were prepared using

appropriate drills with the same taper in a slow-speed

contra-angle handpiece at 1000 rpm. For the 3 mm

post length (group 2 and 4), the apical 3 mm of the

post was cut off to obtain similar dimensions of the post

diameter in the cervical region of all specimens (Fig. 1).

In groups 1 and 2, the size and shape of the bur

corresponded to the FRC post to ensure optimal post fit

(form congruence between post and post space). In

groups 3 and 4, a more extensive post space prepara-

tion was created to simulate missing form-congruence

between post and post space. For that reason, the pilot

drill was shortened by 3 mm apically (group 2) and

6 mm (group 4). Because of the conical shape of the

drill, the diameter of the post space preparation

increases by about 300 lm along the whole length.

This discrepancy between post space and post diameter

results theoretically in a circumferential space width of

150 lm provided that the post is centered in the post

space. The coronal part of the post was reduced in each

group at the same level, i.e. 2.5 mm above the root

canal orifice.

Restorative procedures

Prior to post cementation, the post space was rinsed

with water for 30 s and dried with an air blow for 5 s

and with paper points. Subsequently, all dentine

surfaces were etched with one step (Ultra-etch, 35%

phosphoric acid) for 15 s, rinsed with water spray for

15 s and dried carefully with an air stream for 5 s and

with paper points, leaving the surface slightly moist.

A dual-cure adhesive system (Excite DSC, Ivoclar

Vivadent) was mixed and applied to the sample surface

for 30 s. A gentle air stream was used to evaporate the

dissolution fluid. The FRC posts were cleaned with

alcohol and silanated (Monobond-S, Ivoclar Vivadent)

for 60 s. A dual-curing resin luting material (Multicore

Flow, Ivoclar Vivadent) was mixed and injected into the

prepared root canal with an appropriate tip (C-R

NeedleTubes, Centrix, Shelton, CT, USA). Subsequently,

the post was seated using finger pressure for 10 s.

Excess cement was spread with a brush in a thin layer

Figure 1 Schematic drawing of the post space preparation and the fibre reinforced composite post in the four groups.

Buttel et al. Form-congruence of endodontic posts


so that it covered the occlusal surface of the specimens.

The cement was light-cured (Optilux 500, Demetron/

Kerr, Danbury, CT, USA) for 40 s in an occlusal

direction.

To restore the coronal part of the teeth, direct

composite crowns were built up with the same material

(Multicore Flow). Despite slight differences in the

cervical diameter of the roots, standardized crowns

(4 mm height) were fabricated using transparent

moulds (Pella crowns, Odus, Dietikon, Switzerland)

with anatomically formed occlusal surfaces. Composite

resin was placed free of bubbles in the mould, adapted

to the tooth surface and then light-cured from each side

for 40 s. Finally, the excess composite resin in the

cervical area was removed and the margins of the

restoration were finished using fine diamond burs. In

each specimen, the tip of the post was covered with a

layer of resin composite approximately 1.5 mm in

height.

Mechanical loading

The roots of all specimens were coated with an air-

thinned 0.3-mm layer of polyvinylsiloxane (President

light body, Coltene-Whaledent AG, Altstatten, Switzer-

land) to simulate a periodontal ligament (PDL). The

specimens were fixed with a light-curing composite on

custom-made metallic holders (Provac, Balzers, Liech-

tenstein). The roots were then embedded in self-curing

acrylic resin (Demotec 20, Demotec Siegfried Demel,

Nidderau, Germany) so that the CEJ was situated

approximately 1.5 mm above the simulated bone level

(i.e. the upper margin of the embedding medium). After

embedding, the samples were stored in water until

loading.

All specimens were loaded mechanically at the

centre of the occlusal surface using a computer-

controlled masticator (CoCoM 2, PPK, Zurich, Switzer-

land). Stressing comprised 1.2 million occlusal loads of

49 N at 1.7 Hz obtained by using human cusps.

Simultaneously, thermal stress was applied (3000

thermal cycles between 5/50 �C). These conditions

are believed to simulate approximately 5 years of

clinical service (Krejci et al. 1994).

After thermo-mechanical loading (TML), the fracture

resistance was tested using a universal testing machine

(Zwick, Ulm, Germany). Specimens were fixed in a

metal holder with the long-axis of the roots at an angle

of 45� to the load direction. A tin foil (0.5-mm thick)

was placed between the steel sphere and the crown to

avoid load peaks on the composite resin crown surface.

The linear compressive load was applied (cross-head

speed = 0.5 mm min)1) at the central fissure of the

occlusal surface in the direction of the buccal cusp until

failure.

Statistical analysis

Primary outcome variable was failure during TML

(fatigue testing). Second, loads-to-failure (in N) were

compared when the specimens survived TML. There-

fore, mean values and confidence intervals were

calculated for the nonfailing specimens of each group.

A significant difference between two groups is given

when the confidence intervals do not overlap.

Results

Two specimens, one in group 2 and one in group 4,

were lost because of failures in technical handling. All

remaining teeth and restorations survived TML without

loss of retention or visible fractures and were further

tested for fracture resistance in the universal testing

machine. Mean fracture loads after static loading are

given in Table 1. There was no statistical significant

difference between specimens with 6 mm post length

without form-congruence (group 3) and group 1

(6 mm, form-congruence). Significantly, lower values

were recorded for specimens with a 3-mm short post

(groups 2 and 4). The lowest load values were

Table 1 Failure loads in the four groups

Group n Mean failure load (N) SD (N)

95% Confidence interval for mean

failure load (N)

Lower bound Upper bound

1 (6 mm, form-congruence) 24 393.99A 98.89 352.23 435.75

2 (3 mm, form-congruence) 23 275.47B 75.61 242.77 308.16

3 (6 mm, no form-congruence) 24 408.06A 130.20 353.08 463.04

4 (3 mm, no form-congruence) 23 236.74B 96.27 195.11 278.37

Values exhibiting the same subscript number indicate no significant difference between the groups.



registered in group 4. In the current experimental set-

up, form-congruence had no impact on fracture resis-

tance, irrespective of the post insertion depth.

Discussion

The present study was conducted to evaluate the

influence of form-congruence between post and post

space as well as reduced post length in severely

damaged root filled teeth. It was observed that post fit

did not have a significant impact on fracture resistance,

whilst short post length decreased load values signifi-

cantly, meaning that they fractured more easily.

To gain information about the potential need for a

maximum adaptation of the post to the canal wall, a

lack of form-congruence was created in the current

material by cementing a post in an over-enlarged root

canal. The resulting gap was filled with flowable resin,

which was concurrently used for the crown build-up to

simplify the procedure. This is in accordance with a

recent study suggesting that such materials lead to

better retention than luting cement and therefore

recommending them as alternatives for post cementa-

tion (Ohlmann et al. 2008). The results of the present

study using resin cement clearly demonstrated that

missing form congruence did not impair fracture loads.

In contrast to these findings, Schmage et al. (2005)

reported that only a post closely adapted to the root

canal wall resulted in high retention and prevented

stress peaks. They applied conventional luting material

(zinc phosphate cement) and found that a thin homo-

geneous cement layer, where the film thickness was

<50 lm, was essential to improve post retention.

However, when the composite was used as a luting

material, a mismatch between the diameter of the post

space and that of the post did not impair retention

(Assif & Bleicher 1986, Hagge et al. 2002), even when

shrinkage of the thicker resin cement film resulted in

more stress at the interface between the dentine and

the post (Alster et al. 1997). Perez et al. (2006)

investigated the impact of the resin cement thickness

on the bond strength to the root canal dentine.

Obviously, increased cement thickness did not reduce

the bond strength significantly when FRC posts were

inserted. These findings are in line with those of a

recent study demonstrating that the accuracy of fit

between post and root canal did not influence bond

strength (Perdigao et al. 2007).

With the exception of the two technical failures, in

the current experiment, all the teeth and restorations

survived without loss of post retention or crown

fracture. When loaded to failure, fracture loads in all

groups were found to exceed the chewing forces

normally associated with adults, which ranges from 7

to15 kg (Tortopidis et al. 1998). Teeth with 6-mm deep

post preparations (groups 1 and 3) exhibited similar

failure loads regardless of whether there was form-

congruence between the post and the root canal.

Specimens restored with 3 mm post length with or

without form-congruence (groups 2 and 4) yielded

significantly lower values. Again, the form-congruence

of the FRC-posts had no influence on the load capability

of root filled teeth. During the last decade, the use of

resin composite for direct crowns in root filled teeth has

been recommended only for temporary restorations.

Laboratory investigations of the fracture resistance of

resin composite crowns (with or without endodontic

posts) have, however, yielded promising results, which

suggest that their clinical application is appropriate

(Krejci et al. 1994, Fokkinga et al. 2005). In a 5-year

prospective clinical study on core restorations without

crowns, Creugers et al. (2005a) demonstrated that only

two out of 99 restorations failed. They found that direct

composite build-up restorations exhibited high durabil-

ity and a survival rate similar to that of crowned build-

up restorations in the parallel trial (Creugers et al.

2005b).

To mimic a human periodontium (PDL), the roots of

the tested teeth were covered with a layer of cured

polyvinylsiloxane. The presence of this simulated PDL

was found to significantly affect the results of fracture

testing (Soares et al. 2005). The extracted human teeth

used in the present study were sectioned 1 mm below

the buccal CEJ, thereby removing the enamel com-

pletely. The remaining dentine surface is deemed to

provide poorer bonding characteristics than enamel

(Van Meerbeek et al. 2003). The prepared roots were

provided with posts of different lengths and direct

composite crown build-ups, but no ferrule was

achieved. The advantage of a ferrule is that it generally

facilitates a stabilizing effect by embracing the dentine.

The tooth morphology established here, however,

simulated that of a severely damaged root filled tooth.

This situation is, according to the established clinical

guidelines (Schwartz & Robbins 2004), ideally restored

with a post and core build-up and a custom-made

laboratory crown with circular ferrule (Stankiewicz &

Wilson 2002). In most laboratory studies (Heydecke

et al. 2002, Fokkinga et al. 2006, Salameh et al. 2007),

this clinical recommendation is taken into account and

the tested specimens revealed fracture loads higher

than that in the current investigation. The ferrule



design of crowns is generally considered to be one of the

most important factors to improve the load resistance of

root filled teeth (Sorensen & Engelman 1990, Assif et al.

1993, Isidor et al. 1999, Akkayan 2004, Naumann

et al. 2007). The present results reflect, therefore, the

performance of the post and core alone, without being

confounded by the additional value of a ferruled crown.

In the current material, TML was conducted to

fatigue the samples prior to static loading. A repeatedly

applied load in an aqueous environment simulates

clinical conditions better than static loading alone.

Using this environment, factors such as fatigue stresses

or ageing, which influence the survival of materials can

be taken into account (Naumann et al. 2005). How-

ever, the test designs of laboratory studies can only

partially reflect the clinical situation. Clinically, loading

is a dynamic process and loading forces, frequency and

direction vary greatly. Because of the large number of

other variables involved, including tooth condition,

tooth type, applied restorative procedures and restor-

ative materials used, it is almost impossible to compare

the fracture resistance values obtained in different

laboratory studies. In particular, the most unpredictable

factor is the tooth conditions, which are mainly related

to dentine (Kinney et al. 2003). This is an inherent

drawback associated with the use of human teeth. It has

been reported that testing human teeth results in a large

standard deviation (Krejci et al. 2003), whilst artificially

manufactured teeth are much more consistent (Ottl

et al. 2002). In the present study, a sample size of 24

human teeth was chosen for each group to reduce the

SD and to achieve more reliable results.

The findings of the present study strongly suggest

that excessive post space preparation to maximize post

fit and to reduce the amount of resin cement is not

necessarily required. These results are particularly

encouraging for teeth with oval or long oval root canal

cross-sections. In such cases, not attempting to achieve

a good circumferential post fit helps to preserve inner

dentine and avoids additional weakening of the root.

Further investigations should be conducted to study the

effect of more oval pre-fabricated posts on the load

capability.

Conclusions

Severely damaged root filled teeth restored with FRC

posts and direct resin composite crowns without a

ferrule revealed similar fracture resistance irrespective

of the fit of the post, i.e. irrespective of form-congruence

or no form-congruence. This suggests that post space

preparation and a fitting post are not required to

improve fracture resistance.

Acknowledgements

The authors gratefully acknowledge the Swiss Society

of Odontology (SSO Kuratorium, research project no.

220) for the generous financial support. The authors

would also like to thank Svend Galli, dental technician

and Andres Izquierdo for their valuable help.

References

Akkayan B (2004) An in vitro study evaluating the effect of

ferrule length on fracture resistance of endodontically

treated teeth restored with fiber-reinforced and zirconia

dowel systems. Journal of Prosthetic Dentistry 92, 155–62.

Akkayan B, Gulmez T (2002) Resistance to fracture of

endodontically treated teeth restored with different post

systems. Journal of Prosthetic Dentistry 87, 431–7.

Alster D, Feilzer AJ, de Gee AJ, Davidson CL (1997) Polymer-

ization contraction stress in thin resin composite layers as a

function of layer thickness. Dental Materials 13, 146–50.

Assif D, Bleicher S (1986) Retention of serrated endodontic

posts with a composite luting agent: effect of cement

thickness. Journal of Prosthetic Dentistry 56, 689–91.

Assif D, Bitenski A, Pilo R, Oren E (1993) Effect of post design

on resistance to fracture of endodontically treated teeth with

complete crowns. Journal of Prosthetic Dentistry 69, 36–40.

Creugers NH, Kreulen CM, Fokkinga WA, Mentink AG

(2005a) A 5-year prospective clinical study on core resto-

rations without covering crowns. International Journal of

Prosthodontics 18, 40–1.

Creugers NH, Mentink AG, Fokkinga WA, Kreulen CM

(2005b) 5-year follow-up of a prospective clinical study on

various types of core restorations. International Journal of


Duret B, Reynaud M, Duret F (1990) A new concept of corono-

radicular reconstruction, the Composipost (2). Le Chirurgien-

Dentiste de France 60, 69–77.

Fernandes AS, Dessai GS (2001) Factors affecting the fracture

resistance of post-core reconstructed teeth: a review. Inter-

national Journal of Prosthodontics 14, 355–63.

Fokkinga WA, Kreulen CM, Vallittu PK, Creugers NH (2004)

A structured analysis of in vitro failure loads and failure

modes of fiber, metal, and ceramic post-and-core systems.

International Journal of Prosthodontics 17, 476–82.

Fokkinga WA, Le Bell AM, Kreulen CM, Lassila LV, Vallittu

PK, Creugers NH (2005) Ex vivo fracture resistance of direct

resin composite complete crowns with and without posts on

maxillary premolars. International Endodontic Journal 38,

230–7.

Fokkinga WA, Kreulen CM, Le Bell-Ronnlof AM, Lassila LV,

Vallittu PK, Creugers NH (2006) In vitro fracture behavior of



maxillary premolarswithmetal crowns and several post-and-

core systems. European Journal of Oral Sciences 114, 250–6.

Grandini S, Sapio S, Simonetti M (2003) Use of anatomic post

and core for reconstructing an endodontically treated tooth:

a case report. Journal of Adhesive Dentistry 5, 243–7.

Grandini S, Goracci C, Monticelli F, Borracchini A, Ferrari M

(2005) SEM evaluation of the cement layer thickness after

luting two different posts. Journal of Adhesive Dentistry 7,

235–40.

Gutmann JL (1992) The dentin-root complex: anatomic and

biologic considerations in restoring endodontically treated

teeth. Journal of Prosthetic Dentistry 67, 458–67.

Hagge MS, Wong RD, Lindemuth JS (2002) Effect of dowel

space preparation and composite cement thickness on

retention of a prefabricated dowel. Journal of Prosthodontics

11, 19–24.

Heydecke G, Butz F, Hussein A, Strub JR (2002) Fracture

strength after dynamic loading of endodontically treated

teeth restored with different post-and-core systems. Journal

of Prosthetic Dentistry 87, 438–45.

Isidor F, Odman P, Brondum K (1996) Intermittent loading of

teeth restored using prefabricated carbon fiber posts. Inter-

national Journal of Prosthodontics 9, 131–6.

Isidor F, Brondum K, Ravnholt G (1999) The influence of post

length and crown ferrule length on the resistance to cyclic

loading of bovine teeth with prefabricated titanium posts.

International Journal of Prosthodontics 12, 78–82.

Kinney JH, Marshall SJ, Marshall GW (2003) The mechanical

properties of human dentin: a critical review and re-

evaluation of the dental literature. Critical Reviews in Oral

Biology and Medicine 14, 13–29.

Kishen A, Kumar GV, Chen NN (2004) Stress-strain response

in human dentine: rethinking fracture predilection in

postcore restored teeth. Dental Traumatology 20, 90–100.

Krejci I, Mueller E, Lutz F (1994) Effects of thermocycling and

occlusal force on adhesive composite crowns. Journal of

Dental Research 73, 1228–32.

Krejci I, Duc O, Dietschi D, de Campos E (2003) Marginal

adaptation, retention and fracture resistance of adhesive

composite restorations on devital teeth with and without

posts. Operative Dentistry 28, 127–35.

Lang H, Korkmaz Y, Schneider K, Raab WH (2006) Impact of

endodontic treatments on the rigidity of the root. Journal of

Dental Research 85, 364–8.

Morgano SM (1996) Restoration of pulpless teeth: application

of traditional principles in present and future contexts.

Journal of Prosthetic Dentistry 75, 375–80.

Naumann M, Sterzenbach G, Proschel P (2005) Evaluation of

load testing of postendodontic restorations in vitro: linear

compressive loading, gradual cycling loading and chewing

simulation. Journal of Biomedical Materials Research. Part B,

Applied Biomaterials 74, 829–34.

Naumann M, Preuss A, Frankenberger R (2007) Reinforce-

ment effect of adhesively luted fiber reinforced composite

versus titanium posts. Dental Materials 23, 138–44.

Ohlmann B, Fickenscher F, Dreyhaupt J, Rammelsberg P,

Gabbert O, Schmitter M (2008) The effect of two luting

agents, pretreatment of the post, and pretreatment of the

canal dentin on the retention of fiber-reinforced composite

posts. Journal of Dentistry 36, 87–92.

Ottl P, Hahn L, Lauer H, Fay M (2002) Fracture characteristics

of carbon fibre, ceramic and non-palladium endodontic post

systems at monotonously increasing loads. Journal of Oral

Rehabilitation 29, 175–83.

Perdigao J, Gomes G, Augusto V (2007) The effect of dowel

space on the bond strengths of fiber posts. Journal of


Perez BE, Barbosa SH, Melo RM et al. (2006) Does the

thickness of the resin cement affect the bond strength of a

fiber post to the root dentin? International Journal of


Pirani C, Chersoni S, Foschi F et al. (2005) Does hybridization

of intraradicular dentin really improve fiber post retention in

endodontically treated teeth? Journal of Endodontics 31,

891–4.

Salameh Z, Sorrentino R, Ounsi HF et al. (2007) Effect of

different all-ceramic crown system on fracture resistance

and failure pattern of endodontically treated maxillary

premolars restored with and without glass fiber posts.


Schmage P, Ozcan M, McMullan-Vogel C, Nergiz I (2005)

The fit of tapered posts in root canals luted with zinc

phosphate cement: a histological study. Dental Materials

21, 787–93.

Schwartz RS, Robbins JW (2004) Post placement and resto-

ration of endodontically treated teeth: a literature review.


Soares CJ, Pizi EC, Fonseca RB, Martins LR (2005) Influence of

root embedment material and periodontal ligament simula-

tion on fracture resistance tests. Brazilian Oral Research 19,

11–6.

Sorensen JA, Engelman MJ (1990) Effect of post adaptation on

fracture resistance of endodontically treated teeth. Journal of

Prosthetic Dentistry 64, 419–24.

Sornkul E, Stannard JG (1992) Strength of roots before and

after endodontic treatment and restoration. Journal of


Stankiewicz NR, Wilson PR (2002) The ferrule effect: a

literature review. International Endodontic Journal 35, 575–

81.

Tortopidis D, Lyons MF, Baxendale RH, Gilmour WH (1998)

The variability of bite force measurement between sessions,

in different positions within the dental arch. Journal of Oral

Rehabilitation 25, 681–6.

Trope M, Maltz DO, Tronstad L (1985) Resistance to fracture

of restored endodontically treated teeth. Endodontics and

Dental Traumatology 1, 108–11.

Van Meerbeek B, De Munck J, Yoshida Yea (2003) Adhesion

to enamel and dentin: current status and future challenges.

Operative Dentistry 28, 215–35.



Root canal morphology of mandibular firstpremolars in an Indian population: a laboratorystudy

N. Velmurugan & R. SandhyaDepartment of Conservative Dentistry & Endodontics, Meenakshi Ammal Dental College & Hospital, Chennai, Tamil Nadu, India

Abstract

Velmurugan N, Sandhya R. Root canal morphology of

mandibular first premolars in an Indian population: a laboratory

study. International Endodontic Journal, 42, 54–58, 2009.

Aim To determine the root canal morphology of

mandibular first premolar teeth in an Indian popula-

tion using a decalcification and clearing technique.

Methodology One hundred extracted adult mandib-

ular first premolar teeth were studied following decalci-

fication and clearing. The shape of the canal orifice, root

canal pattern and length of the teeth were determined.

Results The mandibular first premolars were identi-

fied to have a round orifice (38%), oval orifice (44%),

flattened orifice (17%) and C-shaped orifice (1%). The

canal patterns were classified as Type I (72%), Type II

(6%), Type III (3%), Type IV (10%) and Type V (8%)

according to Vertucci’s classification. C-shaped canals

were identified in one tooth (1%). The average length of

the teeth was 21.6 mm. Fourteen per cent of the teeth

had mesial invaginations of the root.

Conclusions Type I canal patterns were the most

frequently occurring in mandibular first premolars

amongst the Indian population. 85.7% of the teeth

with mesial invagination of the root had either two

canals or division of canals.

Keywords: canal orifice, decalcification and clear-

ing, length of the teeth, mandibular first premolar,

mesial invagination.

Received 7 March 2008; accepted 29 September 2008

Introduction

Understanding root canal morphology and its com-

plexity is essential during endodontic therapy. Varia-

tion in the morphology of root canal systems occurs

commonly and can be considered as normal (Cohen &

Hargreaves 2006). Amongst the human permanent

dentition, Brescia (1961) reported that the mandibular

first premolar teeth had the most variable canal

pattern. A study at the University of Washington

assessed the failure rate of nonsurgical root canal

therapy in all teeth. The mandibular first premolar had

the highest failure rate and this may be attributed to

the frequent variations in the root canal morphology

and the inability to access extra canals (Ingle & Taintor

1985).

It is a well known fact that the root canal system

varies with race (Trope et al. 1986, Ahmed et al.

2007), and gender (Sert & Bayirli 2004). Earlier studies

on root canal systems were completed most commonly

on teeth from Caucasian populations. Similar studies

amongst the Indian population are rare (Reuben et al.

2008). The aim of this study was to determine the root

canal morphology of mandibular first premolar teeth in

an Indian population using a decalcification and

clearing method.


One hundred extracted human adult mandibular first

premolar teeth from an Indian population were col-

lected. The age and gender of the patients were not

known. Teeth with deep caries, metallic restorations,

Correspondence: Dr N. Velmurugan, # 2, 95th Street, 21st

Avenue, Ashok Nagar, Chennai 600083, Tamil Nadu, India

(Tel.: 9840164167; fax: 044 2378 1631; e-mail: vel9911

@yahoo.com).

doi:10.1111/j.1365-2591.2008.01494.x


fracture, incompletely formed roots and those which

were root filled were not included. Handling of the teeth

was carried out according to Occupational Safety and

Health Administration guidelines and regulations.

The teeth were preserved in 10% formalin (Western

India Chemical, Udupi District, Karnataka, India). All

attached soft tissue and calculus were removed using

an ultrasonic scaler. The length of the teeth was

measured using vernier caliper from the tip of the

crown to the apex of the root. In case of a curved root,

tangents were drawn to the curved portions of the

tooth. The length was then measured by connecting

the points of tangency.

The teeth were decalcified and rendered transparent

using the technique reported by Robertson et al. (1980)

to obtain a 3D view of the root canal system. Access

cavities were prepared using a round bur (No. 2 round

bur) and the shape of the canal orifice was observed

with the naked eye. Following this, the teeth were

placed in 3% sodium hypochlorite (Merck Limited,

Mumbai, Maharashtra, India) for 48 h. The teeth were

agitated manually to ensure complete removal of the

pulp tissue. The teeth were then washed in running

water for 2 h and then transferred to 5% nitric acid

(Merck Limited) for decalcification. The teeth were

placed in acid for 72 h, with the acid being changed

every 24 h and stirred once every 8 h. The end-point of

decalcification was determined by taking a radiograph

of three sample teeth, which showed uniform decalci-

fication of the teeth. The teeth were then washed in

running water and dehydrated using ascending grades

(70%, 80%, 90% and 100%) of isopropyl alcohol

(Leonid Chemicals Pvt Ltd, Bangalore, Karnataka,

India) for 2 days. Finally, they were rendered trans-

parent by immersion in methyl salicylate (Sipali

Chemicals, Chennai, Tamilnadu, India) and an oil-

based dye was injected into the access cavity. The

anatomy of the root canal was observed and classified

based on the Vertucci’s classification (Vertucci 1984).

The supplementary canals present at the apical third

were grouped as accessory canals and those in the

middle third as lateral canals. Fourteen teeth in the

study had invagination of the root surface on its mesial

aspect. These teeth were analysed to check for any

specific variations of the canal anatomy that could be

associated with this feature.

Results

Canal orifice

The shape of the canal orifices were round in 38% of

the teeth, oval in 44% of the teeth, flattened ribbon

shaped in 17% of the teeth and C-shaped in 1% of the

teeth. Two canal orifices were seen in 2% of the teeth.

Canal type

Amongst the 100 mandibular first premolar teeth,

72% had a Type I canal pattern (Fig. 1a) with Type II,

Type III, Type IV and Type V canals being identified in

6%, 3%, 10% and 8% of the teeth respectively

(Fig. 1b–e, Table 1). One tooth had Category III

c-shaped canal (1%) (Melton et al. 1991). Lateral

canals were observed in 4% of the samples and

another 4% of the samples had accessory canals.

Intercanal communication was identified in only one

tooth sample (1%).

Position of the apical foramen

Amongst the teeth with a single canal at the apex

(n = 82), the apical foramen was located at the apex of

the root in 83% teeth, 0.5 mm from the apex in 6%

a) (b) (c) (d) (e)

Figure 1 Various canal patterns in mandibular first premolars. (a) Type I, (b) Type II, (c) Type III, (d) Type IV, (e) Type V.

Velmurugan & Sandhya Root canal morphology


teeth, 1 mm from the apex in 9.7% teeth and 2 mm

from the apex in 1.2% teeth.

Length of teeth

The longest tooth in this study was 25.2 mm and the

shortest was 17.7 mm. The average length of the

mandibular first premolar teeth was 21.6 mm, the

median and mode were 21.3 mm.

Mesial invagination of the teeth

Mesial invagination of the root was found in 14% of the

teeth. Amongst them, 7 teeth had Type IV canal

pattern; 3 had Type V canal pattern and 2 had Type I

canal pattern (Fig. 2 Table 2). One tooth with Type II

and one tooth with Type III canal pattern were also

identified. 85.7% of the teeth with mesial invagination

had either two canals or division of canals. The mean

distance from the cusp tip to the point of initiation of

invagination was 14.6 mm.

Discussion

This study analysed the canal morphology of mandib-

ular first premolar teeth amongst an Indian population

using a decalcification and clearing technique. Previous

studies report a high occurrence of Type I canal pattern

(Vertucci 1984). Studies on root canal anatomy have

Table 1 Pattern and percentage of canals

Type of

canal

Canal

pattern

% of

Occurrence

(n = 100)

Type Ia 1 72

Type IIa 2-1 6

Type IIIa 1-2-1 3

Type IVa 2 10

Type Va 1-2 8

Type VIa 2-1-2 0

Type VIIa 1-2-1-2 0

Type VIIIa 3 0

Cb 1-3-1 1

aVertucci (1984).bMelton et al. (1991).

(a)

(a1)

(b)

(b1)

(c)

(c1)

Figure 2 Mandibular first premolars with mesial invagination of the root (a, b, c) and their root canal patterns (a1, b1, c1).

Root canal morphology Velmurugan & Sandhya


been conducted using methods, such as radiography

(Pineda & Kuttler 1972, Willershausen et al. 2006),

decalcification and clearing (Caliskan et al. 1995,

Rwenyonyi et al. 2007), direct observation with micro-

scope (Sempire & Hartwell 2000), 3D reconstruction

(Mikrogeorgis et al. 1999), computed tomography

(Robinson et al. 2002, Reuben et al. 2008) and mac-

roscopic sections (Baisden et al. 1992, Lu et al. 2006).

It has been mentioned that the most detailed informa-

tion can be obtained by demineralization and clearing

technique (Vertucci 1984). Moreover, it is simple,

acceptable and an inexpensive procedure (Rwenyonyi

et al. 2007).

The most prevalent canal pattern in the present

study was Type I occurring in 72% of the mandibular

first premolars (Fig. 1a, Table 1). In an earlier study

(Vertucci 1984) in Caucasian population, the preva-

lence was 70%, whereas other studies have reported a

Type I canal pattern in 67.2% to 86.3% of teeth(Zillich

& Dowson 1973, Trope et al. 1986). A Type II canal

was encountered in 6% of the samples and Type V in

8% of the samples (Fig. 1, Table 1). Vertucci (1984) did

not report any Type II canal patterns, but 24% of the

teeth in his study had a Type V canal pattern. These

variations may be attributed to the racial or genetic

factors. Vertucci (1984) reported the occurrence of

C-shaped canal in 0.5% of the samples, whereas in the

present study, it was identified in one tooth (1%)

(Table 1). Melton et al. (1991) classified C-shaped

canals into three types. The C-shaped canal identified

in this study was Category III sub division I, where the

canal divided into three in the middle third and

reunited at the apical region to exit through one

foramen.

A previous study reported the average length of

mandibular first premolar teeth to be 21.6 mm (Cohen

& Hargreaves 2006). The average length of the teeth in

the present study was also found to be 21.6 mm. In this

study there were 14 teeth with a mesial invagination of

the root. The point of initiation and the depth of the

invagination varied. According to Ash (1999), these

are deep developmental grooves found on the mesial

surface of the root. Radiographic studies on canal

anatomy have not reported on mesial invaginations of

roots as it is impossible to identify its presence in

clinical radiographs. The only in vivo study that

reported their occurrence using spiral computed

tomography concluded that 15% of mandibular pre-

molars had invagination (Robinson et al. 2002).

According to that study, the mesial invagination gave

a false radiographic line that one can mistake for an

extra canal. In this study, amongst the 14 teeth having

mesial invagination of the root, 8 teeth had two canals,

4 teeth had bifurcation of canal and 2 teeth had a

single canal. The teeth with two canals (Type II and

Type IV) had the mesial invagination initiating from

the cervical half of the tooth root. The teeth with canal

bifurcation (Type III and Type V) and single canal

(Type I) had the invagination in the apical half of the

root, with the single canal specimen having the

invagination apically. In these teeth, the lingual canal

after bifurcation was smaller in diameter when com-

pared with the buccal canal. The location of the canal

bifurcation varied in accordance with the location of

the point of initiation of invagination. There seems to

be some anatomical correlation between mesial invag-

ination of the root and canal pattern which requires

further analysis.

Conclusion

A Type I canal pattern was found to be the most

prevalent in mandibular first premolar teeth amongst

this Indian population. More than one canal was

commonly found in the teeth with mesial invagination

of the root.

References

Ahmed HA, Abu-bakr NH, Yahia NA, Ibrahim YE (2007) Root

canal morphology of permanent mandibular molars in a

Sudanese population. International Endodontic Journal 40,

766–71.

Ash M (1999) Wheeler’s Dental Anatomy, Physiology and

Occlusion, 7th edn. Philadelphia: W.B. Saunders Company,

pp. 228–9.

Baisden MK, Kulild JC, Weller RN (1992) Root canal config-

uration of the mandibular first premolar. Journal of End-

odontics 18, 505–8.

Table 2 Mesial invagination of the teeth

Canal

type

No. of

teeth

Average point

of initiation of

invagination (mm)a

Type I 2 16.5

Type II 1 14.4

Type III 1 15.3

Type IV 7 13.8

Type V 3 15.1

aPoint of initiation of invagination measured from the cusp tip.

Average: 14.6 mm; median: 14.4 mm; mode: 15.3 mm; vari-

ance: 2.5; SD: 1.59.

Velmurugan & Sandhya Root canal morphology


Brescia NJ (1961) Applied Dental Anatomy. St. Louis: CV Mosby

Co., pp. 46–8.

Caliskan MK, Pehlivan Y, Sepetcioglu F, Turkun M, Tuncer SS

(1995) Root canal morphology of human permanent teeth

in Turkish population. Journal of Endodontics 21, 200–4.

Cohen S, Hargreaves KM (2006) Pathways of the Pulp, 9th edn.

St. Louis: Mosby – Elsevier, pp. 216–7.

Ingle JI, Taintor JF (1985) Endodontics, 3rd edn. Philadelphia:

Lea & Febiger, pp. 27–52.

Lu TY, Yang SF, Pai SF (2006) Complicated root canal

morphology of mandibular first premolar in a Chinese

population using the cross section method. Journal of


Melton DC, Krell KV, Fuller MW (1991) Anatomical and

histological features of C- shaped canals in mandibular

second molars. Journal of Endodontics 17, 384–8.

Mikrogeorgis G, Lyroudia KL, Nikopoulos N, Pitas I, Molyvdas

I, Lambrianidis TH (1999) 3D computer-aided reconstruc-

tion of six teeth with morphological abnormalities. Interna-


Pineda F, Kuttler Y (1972) Mesiodistal and buccolingual

roentgenographic investigation of 7,275 root canals. Oral

Surgery Oral Medicine and Oral Pathology 33, 101–10.

Reuben J, Velmurugan N, Kandaswamy D (2008) The

evaluation of root canal morphology of the mandibular first

molar in an Indian population using spiral computed

tomography scan: an in vitro study. Journal of Endodontics

34, 121–249.

Robertson D, Leeb J, Mckee M, Brewer E (1980) A clearing

technique for the study of root canal systems. Journal of


Robinson S, Czerny C, Gahleitner A, Bernhart T, Kainberger

FM (2002) Dental CT evaluation of mandibular first

premolar root configurations and canal variations. Oral

Surgery Oral Medicine Oral Pathology Oral Radiology and


Rwenyonyi CM, Kutesa AM, Muwazi LM, Buwembo W (2007)

Root and canal morphology of maxillary first and second

permanent molar teeth in a Ugandan population. Interna-


Sempire HN, Hartwell GR (2000) Frequency of second

mesiobuccal canals in maxillary molars as determined by

use of an operating microscope: a clinical study. Journal of


Sert S, Bayirli GS (2004) Evaluation of the root canal

configurations of the mandibular and maxillary permanent

teeth by gender in the Turkish population. Journal of


Trope M, Elfenbein L, Tronstad L (1986) Mandibular premo-

lars with more than one root canal in different race groups.


Vertucci FJ (1984) Root canal anatomy of the human

permanent teeth. Oral Surgery Oral Medicine Oral Pathology

Oral Radiology and Endodontics 58, 589–99.

Willershausen B, Tekyatan H, Kasaj A, Marroquin BB (2006)

Roentgenographic in vitro investigation of frequency and

location of curvatures in human maxillary premolars.


Zillich R, Dowson J (1973) Root canal morphology of

mandibular first and second premolars. Oral Surgery Oral

Medicine Oral Pathology 36, 738–44.

Root canal morphology Velmurugan & Sandhya


Necrotic pulp tissue dissolution by passiveultrasonic irrigation in simulated accessory canals:impact of canal location and angulation

A. Al-Jadaa, F. Paque, T. Attin & M. ZehnderDepartment of Preventive Dentistry, Periodontology and Cariology, University of Zurich Center of Dental Medicine, Zurich,

Switzerland

Abstract

Al-Jadaa A, Paque F, Attin T, Zehnder M. Necrotic pulp

tissue dissolution by passive ultrasonic irrigation in simulated

accessory canals: impact of canal location and angulation.

International Endodontic Journal, 42, 59–65, 2009.

Aim To evaluate whether passive ultrasonic irrigation

(PUI) of 2.5% NaOCl would dissolve necrotic pulp tissue

from simulated accessory root canals (SACs) better

than passive placement of the irrigant, when temper-

ature was equilibrated between the two treatments.

Methodology Transparent root canal models

(n = 6) were made from epoxy resin. SACs of 0.2 mm

diameter were placed at defined angles and positions in

the mid-canal and apical area. SACs were filled with

necrotic bovine pulp tissue. PUI was performed five

times for 1 min each with irrigant replenishment after

every minute. Main canal temperature was measured

after each minute, and a digital photograph was taken.

In control experiments, mock treatments were per-

formed with the same set-up without activation of the

file using heated NaOCl to mimic the temperature

created by PUI. Experiments were repeated five times.

Digital photographs were analysed for the distance of

dissolved tissue into the SACs in mm. Overall compar-

ison (sum of dissolved tissue from all five accessory

canals) between treatments was performed using

paired t-test. Differences between SAC angulation and

position after PUI were investigated using anova/

Bonferroni (alpha < 0.05).

Results Passive ultrasonic irrigation caused a rise in

irrigant temperature in the main canal to

53.5 ± 2.7 �C after the fifth minute. PUI dissolved a

total of 6.4 ± 2.1 mm, mock treatment controlled for

heat: 1.4 ± 0.6 mm (P < 0.05). No significant influ-

ence of SAC position or angulation was found.

Conclusions Passive ultrasonic irrigation promotes

positive tissue-dissolving effects beyond a rise in

irrigant temperature.

Keywords: sodium hypochlorite, passive ultrasonic

irrigation.

Received 30 July 2008; accepted 3 October 2008

Introduction

Disinfection and debridement of root canals is an

important aspect of endodontic treatment. Based on the

fact that mechanical preparation alone cannot fully

achieve this aim (Bystrom & Sundqvist 1981), the

chemo-mechanical principle using topically applied

substances during and after instrumentation was

established. In this context, the correct choice of the

chemicals to be used and their ideal mode of application

are of interest. Sodium hypochlorite is the root canal

irrigant of choice for many practitioners, as it dissolves

necrotic tissue (Naenni et al. 2004) and has a superior

antimicrobial effect compared with most other disin-

fectants that have been used in the root canal system

(Vianna et al. 2006). It has been shown that the local

efficacy of hypochlorite preparations can be improved

Correspondence: Matthias Zehnder, PD Dr med dent PhD,

Department of Preventive Dentistry, Periodontology and

Cariology, University of Zurich Center for Dental Medicine,

Plattenstrasse 11, CH 8032 Zurich, Switzerland (Tel.:

+41 44 632 8610; fax: +41 44 634 4308; e-mail: matthias.

[email protected]).

doi:10.1111/j.1365-2591.2008.01497.x


by heating the solution to be applied (Sirtes et al.

2005). Alternatively, the irrigant can be activated

mechanically. Amongst the mechanical methods for

irrigant activation, passive ultrasonic irrigation (PUI) is

probably the most established method (van der Sluis

et al. 2007).

Ultrasound was first introduced to endodontics in

1957 for mechanical root canal and root-end prepara-

tion (Richman 1957). Later, it was realized that

ultrasonic activation could be beneficial in enhancing

the efficacy of irrigants in the root canal (Martin 1976,

Martin & Cunningham 1985). The main effects in this

context are (transitional) cavitation and streaming

(Walmsley 1987). Both phenomena are well known to

enhance the effectiveness of antiseptics, especially

sodium hypochlorite (Martin & Cunningham 1985,

Blume & Neis 2005). Whilst streaming undoubtedly

occurs, it is unclear whether cavitation actually occurs

in the root canal system (Ahmad et al. 1987, Lumley

et al. 1988). A third, often overlooked, effect of the

application of ultrasonic energy in the root canal is the

general increase in irrigant temperature (Cunningham

et al. 1982, Cameron 1988).

Researchers have extensively studied the influence of

ultrasonic irrigant activation on the appearance of root

canal walls as observed by scanning electron micros-

copy (Ahmad et al. 1987, Abbott et al. 1991). Others

used a scoring model of the stained organic debris and

smear layer (Cheung & Stock 1993). It was found that

ultrasonic activation increases the debridement activity

of sodium hypochlorite (Cameron 1987). Using artifi-

cially prepared grooves filled with dentine debris in the

walls of human root canals as well as in artificial

canals, it has been shown that PUI has the potential to

remove debris from canal extensions and irregularities

(van der Sluis et al. 2005). It was also shown in situ

that the soft tissue debridement of sodium hypochlorite

is greatly enhanced by ultrasonic activation in the

isthmus areas of human mandibular molars (Burleson

et al. 2007). However, until now the impact of PUI on

accessory canals is still unclear because of the lack of

studies with such observations. It has been shown that

clinically, these areas are especially difficult to clean

(Nair et al. 2005). The lack of studies on irrigant action

in lateral or accessory canals can be related to the

difficulty in carrying out such investigations on natural

teeth, as the accessory canal position and status before

treatment are difficult to determine. Consequently,

there appears to be a need for standardized models

simulating accessory canals with multiple controlled

variables yielding repeatable results. The aim of this

study was to establish a model especially tailored for

this purpose.


Fabrication of model

A suitable model that would allow the observation and

direct quantitative measurement of pulp tissue before

and after irrigation was not available. A transparent

model was prepared using a wax mould that was filled

with epoxy resin (Stycast, Emerson & Cuming, Wester-

lo, Belgium). To ensure reproducibility of the model, a

sheet of paper with a drawing representing the main

canal, position and angulation of accessory canals was

used as reference to assemble the parts in the proper

position using super glue before transferring them to a

box made of pink plate wax with a dimension of 30, 20

and 15 mm length, width and height, respectively

(Fig. 1a,b). The main canal was simulated using a

D-size finger spreader (Dentsply Maillefer, Ballaigues,

Switzerland). This instrument had a length of 25 mm,

a tip diameter of 0.35 mm, and a 0.06 taper (Briseno

Marroquın et al. 2001). Accessory canals were created

by 0.2-mm stainless steel wires (Fig. 1b,c). The length

of the canal was determined by allowing 5 mm of the

wire to extrude from a 22-gauge needle (Ultradent

Products, Inc. South Jordan, UT, USA), The needle was

used later to carry the necrotic pulp tissue and apply it

into the canal by means of injection. A pair of canals

were placed at distances of 1 mm and 9 mm from the

main canal apex opposing each other, one of these was

made perpendicular to the main canal, the other

created at a 45� angle with the apical extension of

the main canal. In addition, an accessory canal that

continued in the direction of the main canal (180�) was

created. A millimetric paper scale was placed parallel to

the long access of each simulated accessory canal to

ensure a precise measurement of the length of tissue

dissolution. Eight models were fabricated to be used in

the study. Before any of the models were used,

continuity of simulated accessory canals with the main

canal was ensured by introducing a 0.2-mm wire inside

each accessory canal until it appeared in the main

canal. Finally, a simulated pulp chamber and reservoir

for the passively placed irrigant was created using a

rubber tube with a length of 7 mm and 3 mm internal

diameter, which was glued over the main canal

entrance. This reservoir ensured that the whole canal

remained filled with irrigant after the passive ultrasonic

activation procedure described below. A model ready to

PUI in accessory canals Al-Jadaa et al.


be filled with necrotic pulp tissue is depicted in Fig. 1,

panel d.

Bovine pulp tissue preparation

The accessory canals of seven models were filled with

bovine pulp tissue. The tissue was obtained from bovine

anterior teeth of animals that were raised and slaugh-

tered for food production according to the Swiss

standards of animal welfare. Consequently, this study

was not considered an animal study and the internal

review board had no objections to the current protocol.

Pulps were extirpated after decoronation of the teeth

and then frozen at )20 �C. Frozen tissue was thawed,

dried with paper tissues, and then each piece was

immersed in liquid nitrogen to achieve a solid dry

material. Subsequently, tissue was transformed into

fine particles using a scalpel to scratch the hard surface.

Sometimes it was required to re-immerse the piece into

liquid nitrogen several times to maintain its solid

consistency. When a sufficient amount of tissue

was prepared, a 22-gauge needle (Ultradent Products)

was used to aspirate part of it and then the needle was

inserted in its place in the model until it reached the

outer end of the simulated accessory canal. The tissue

was injected in the accessory canal until part of it

(a) (b)

(c) (d)

Figure 1 Preparation of an epoxy resin

model used in this study: (a) template

to ensure similar simulated accessory

canal position and angulation between

the models; (b) positioning of the finger

spreader and the wires; (c) mould

made of pink wax filled with epoxy resin;

(d) finished model.

Al-Jadaa et al. PUI in accessory canals


extruded into the main canal. Excess tissue was placed

in the wide entrance of the carrying needle to obtain

a passive closure simulating a pathosis rather than

a tight seal of the simulated accessory canals. This

procedure was repeated in all the five simulated

accessory canals in each model.

The models were re-filled for the control experiments

with heated NaOCl and NaOCl at room temperature

(see below) after removing the old tissue from accessory

canals and extensive rinsing with tap water.

Control experiments on temperature

It is well known that ultrasonic irrigant activation is

associated with heat generation (Cunningham et al.

1982, Cameron 1988). An increase in temperature can

enhance the efficacy of NaOCl (Sirtes et al. 2005). To

discern between pure temperature and other ultrasonic

effects on NaOCl, the temperature associated with PUI

in the current model was determined. A preliminary

study was carried out using one of the models

fabricated for the study. The temperature was recorded

after each 1 min of activation and also after each flush

with 1 mL 2.5% (wt/vol) NaOCl using a thin couple

wire connected to a calibrated temperature measuring

device (Testo Term 9010, Lenzkirch, Germany). This

procedure was carried out over 5 min and repeated

thrice. After the intracanal temperature created by PUI

in the current set-up was known, the irrigant temper-

ature to be used in the second part of the study was

determined by trial. The irrigant was heated by placing

the irrigation syringe inside a water bath and the

temperature was measured after each irrigation by

1 mL of 2.5% NaOCl and after 1 min of irrigation. After

that 1 min the syringe was returned to the water bath

to ensure a stable temperature. The irrigant tempera-

ture inside the syringe was measured by introducing

the couple wire through its opening just before

irrigation. The temperature was raised gradually until

the suitable temperature inside the canal was achieved.

This procedure was repeated thrice.

Main experiment

The model was held on a cone especially designed to

direct light through it to have a contrast facilitating the

interpretation of results and to prevent artefacts caused

by over-exposure of light. Halogen light (Intralux

4000-1, Volpi AG, Schlieren, Switzerland) was intro-

duced from behind the model and through the cone. An

initial photograph using a 10-megapixel camera

(Nikon D200, Tokyo, Japan) mounted on a stand in

front of the model was taken to ensure the complete

filling of the simulated accessory canals with pulp tissue

and to allow comparison later on. The irrigation

protocol was as follows: 1 mL of 2.5% NaOCl at room

temperature was introduced to full canal length by a

long irrigation needle with 30-gauge diameter (Max-i-

Probe, Hawe Neos, Bioggio, Switzerland). Care was

exercised that the opening at the needle tip was not

directed towards the accessory canals directly. An

ultrasonic device (EMS 400, EMS, Nyon, Switzerland)

with its power set at the ¼ of the scale, with an

ultrasonic stainless steel K-type file size 15 (Endoso-

nore, Dentsply Maillefer) mounted on an ultrasonic

adaptor (Piezon, 90� Endo File Holder, EMS) was used

to activate the irrigant in the canal with an up and

down motion by hand at a ratio of 10 mm s)1 to the

full length of the canal minus half a millimeter, for

1 min. Subsequently, a photo was taken and the main

canal was irrigated with 1 mL of sodium hypochlorite

at room temperature. The same procedure was

repeated every minute for 5 min. At the end of the

fifth minute, the temperature inside the canal was

measured to ensure that the ultrasonic file was active.

The ultrasonic file was replaced for each model to avoid

fracture, whilst the ultrasonic adaptor was replaced

after two models. This protocol was carried out on the

seven models. In the control experiments, the models

were refilled with tissues as described before and the

same procedure was carried out except for the NaOCl

temperature which was 68–69 �C in the second

experiment and at room temperature the third time.

The file was introduced in the canal without ultrasonic

activation in these two experiments. The experiment

for the second and third parts was carried out only on

six models because one of the models was lost because

of a fractured file in the first part. Results from that

model were discarded.

Data generation and analysis

Data from the temperature experiments are presented

as means and standard deviations (n = 3).

The photos were analyzed using the ImageJ program

(nih.gov; National Institute of Health, Bethesda, MD,

USA). The outcome variable assessed here was distance

of tissue dissolution in simulated accessory canal,

measured from the canal entrance to the closest

tissue-irrigant interface. Measurements were performed

by one operator, who was tested for his accuracy by

analysing the same images ten times after different



intervals. The error of the individual measurement was

< 0.05 mm. Consequently, data pertaining to tissue

dissolution were rounded to 0.1 mm. To compare

overall tissue dissolution at room temperature with the

corresponding values obtained by PUI and in the

temperature-controlled experiments, the sums of dis-

tances of tissue dissolution in all accessory canals per

model were averaged for each mode (n = 6) and

compared by a paired t-test. To compare the impact of

accessory canal position and angulation on tissue

dissolution by PUI, mean values per simulated accessory

canal were compared by one-way analysis of variance

(anova). Bonferroni’s correction was applied for multi-

ple testing. The alpha-type error was set at 0.05.

Results

Temperature

Passive ultrasonic irrigation caused a rise in hypochlo-

rite temperature in the main canal to 53.5 ± 2.7 �Cafter the fifth min (Fig. 2). For the temperature-control

experiment, the suitable irrigant temperature in the

syringe was found to be 68–69 �C, which was achieved

by placing the 5-mL irrigation syringe in a water bath

of 75 �C for 5 min. This resulted in an overall

temperature in the canal that was similar to the one

observed with PUI (Fig. 2).

One of the observations, which might affect the

clinical usability of PUI, was that after multiple usage of

the ultrasonic adaptor (usually after 12–14 min of

activation), the temperature suddenly dropped, indicat-

ing a loss of ultrasonic energy transmitted to the irrigant

in the canal. After multiple trials and by exclusion it was

found that the rubber ring between the two parts of the

ultrasonic adaptor wore out so that there was less

activation of the ultrasonic file. This observation

necessitated a regular replacement of the adaptor. As

an extra precaution the temperature wasmeasured after

the fifth and final minute of PUI in each individual model

as an indicator of the ultrasonic activity inside the canal.

Tissue dissolution

The mean sums of dissolved tissue from simulated

accessory canals after 5 min of PUI or the mock

treatments were: PUI: 6.4 ± 2.1 mm, mock treatment

at room temperature: 0.8 ± 0.3 mm, and mock treat-

ment controlled for heat: 1.4 ± 0.6 mm. The difference

between the heated irrigant and the counterpart

administered at room temperature was not significant

at the 0.05 level, whilst there was a significant

(P < 0.05) difference between both these treatments

and PUI, indicating a clear PUI effect.

When the influence of simulated accessory canal

position and angulation on tissue dissolution by PUI

was studied (Table 1), it was noted that regardless of

accessory canal position or angulation, a plateau was

reached after the third minute of activation. Further-

more, there was no significant difference in tissue

dissolution between different simulated accessory

canals at any time.

Discussion

The current study showed a positive effect of PUI in

conjunction with a sodium hypochlorite irrigant on

pulp tissue dissolution from simulated accessory canals

in an epoxy resin model. This effect was not explained

by a simple rise in overall irrigant temperature.

The current study is limited by the fact that epoxy

resin is a completely different material from human

dentine, and direct clinical conclusions can therefore

not be drawn from the results presented here. Further-

Figure 2 Temperatures (�C) measured in

the simulated main canal after passive

ultrasonic irrigation (blue) and during

the mock treatment with a heated

sodium hypochlorite solution (red) over

time. Dots indicate means, error bars

standard deviations (n = 3).



more, the simulated main canal in the current model

was straight. This type of anatomy is rarely encountered

in natural teeth. However, the aim of this study was to

discern between mere temperature and other PUI effects

in the cleansing of accessory canals. For this purpose,

the model appeared adequate. However, despite the

standardization of the models that were used, data

variation pertaining to the distance of dissolved tissue in

simulated accessory canals was still relatively large as

indicated by the high standard deviations (Table 1).

This can be explained by the difficulty in obtaining

completely homogenous and standardized fills of

necrotic tissue in these thin canals. On the other hand,

the density of necrotic tissue in infected natural

accessory canals might also vary. It is a common

observation when dealing with natural tissues such as

the bovine pulps that were used in the current inves-

tigation that outcomes vary. In addition, because the

ultrasonic tip was guided by hand, it was impossible to

control where it touched the canal wall, which may also

have contributed to the variance in outcome. A further

limitation of this study is the fact that the average width

of accessory canals is not known or published (De Deus

1975). However, based on our own observations on

micro-computer tomographies of human teeth, 200 lmappeared to be a fair approximation.

The temperature that was measured in the current

study was somewhat higher than that measured in

natural teeth, which may be because of the fact that

thermal transducing properties of dentine differ from

those of epoxy resin (Brown et al. 1970) and also

because of the potential cooling effect of the blood

circulation around natural teeth. Using PUI with

intermittent flushes, temperatures of up to 45 �C were

measured in root canals of natural teeth after 30 s of

ultrasonic irrigant activation (Cameron 1988). Consid-

ering the shorter activation times, there appears to be

little variance between these published data and the

current results. However, other researchers found the

temperature rise in the root canal promoted by PUI to

be minimal (Ahmad 1990). However, a root canals

were widened to an ISO-size 80 in that study, and there

was continuous flow of irrigant during ultrasonic

activation which might explain the differences.

The exact mechanism by which ultrasonic hypo-

chlorite activation can affect the tissue in accessory

canals is still unclear. One hypothetical mechanism is

the collapse of bubbles during transient cavitation that

produces a pressure-vacuum effect, which sucks the

canal content to the inside rather than pushing it

further in the canal. This will be followed by diffusion of

the irrigant in the main canal to substitute the space

created (Martin & Cunningham 1985). Another possi-

bility is that the streaming around the activated file

because of the cohesion between fluid particles inside

the accessory canal and the irrigant in the main canal

sucks the content of the accessory canals into the main

canal with fluid flow toward the main canal (Ahmad

et al. 1992). The third possibility is a local temperature

effect because of the collapse of bubbles during transi-

tional cavitation. It has been shown that locally, the

temperature can reach up to 5000 �C with heating and

cooling rates greater than 109 K/s during cavitation

(Suslick 1990). Consequently, a great part of the

ultrasonic effect may still be thermal, but just not

measurable by assessing the overall irrigant tempera-

ture. However, it is still unclear at this point whether

transient cavitation occurs in the root canal. Based on

preliminary observations with dye solutions of different

colours in the model described here, it was noted that

little streaming occurred in the apical area, especially in

the simulated accessory canal at 180� at the apical endof the main canal (not shown). Nevertheless, tissue

dissolution was similar regardless of accessory canal

position or angulation in the current study. Conse-

quently, it may be so that cavitation was, at least in

part, responsible for the observed phenomenon of tissue

dissolution by PUI. This again highlights what has been

pointed out more than 20 years ago, namely that

further studies are required to elucidate the phenomena

behind ultrasonic effects that might or might not occur

in the root canal.

Table 1 Distance in mm of dissolved tissue as measured from the simulated accessory canal entrance after passive ultrasonic

irrigation (means and standard deviations, n = 6)

Time 90�, mid-canal 45�, mid-canal 90�, apex 45�, apex 180�, apex

1st min 0.2 ± 0.3 0.1 ± 0.2 0.0 ± 0.0 0.4 ± 0.6 0.1 ± 0.1

2nd min 1.1 ± 0.5 0.8 ± 0.6a 0.9 ± 0.3 1.3 ± 0.8 0.5 ± 0.6

3rd min 1.4 ± 0.5 1.0 ± 0.6 1.1 ± 0.3 1.5 ± 0.9 0.7 ± 0.7

4th min 1.4 ± 0.5 1.1 ± 0.6 1.2 ± 0.3 1.6 ± 0.8 0.8 ± 0.8

5th min 1.5 ± 0.6 1.2 ± 0.6 1.3 ± 0.3 1.7 ± 0.8 0.9 ± 0.9

No statistically significant differences were found between canals at any given time (P > 0.05, anova, Bonferroni).



One further observation that was made during the

current study was that because of the high corrosive

potential of hypochlorite and the heat that is generated

during ultrasonic activation, material wear out oc-

curred rapidly. Initially, noncutting nickel-titanium tips

were used, but these fractured so frequently that it was

decided to use the cheaper stainless-steel files. Results

between the two types of instruments were similar

(data not shown).

Conclusions

• A model allowing the quantitative assessment of

necrotic pulp tissue dissolution in simulated accessory

canals was presented.

• The temperature generated in the main canal of this

model by passive ultrasonic activation of a 2.5% NaOCl

solution was over 50 �C.• This rise in overall temperature could not be

responsible for the effectiveness of PUI.

• Tissue dissolution by PUI was irrespective of simu-

lated accessory canal position or angulation.

References

Abbott PV, Heijkoop PS, Cardaci SC, Hume WR, Heithersay GS

(1991) An SEM study of the effects of different irrigation

sequences and ultrasonics. International Endodontic Journal

24, 308–16.

Ahmad M (1990) Measurements of temperature generated by

ultrasonic file in vitro. Endodontics and Dental Traumatology

6, 230–1.

Ahmad M, Pitt Ford TR, Crum LA (1987) Ultrasonic debride-

ment of root canals: an insight into the mechanisms

involved. Journal of Endodontics 13, 93–101.

Ahmad M, Roy RA, Kamarudin AG (1992) Observations of

acoustic streaming fields around an oscillating ultrasonic

file. Endodontics and Dental Traumatology 8, 189–94.

Blume T, Neis U (2005) Improving chlorine disinfection of

wastewater by ultrasound application. Water Science and

Technology 52, 139–44.

Briseno Marroquın B, Wolter D, Willershausen-Zonnchen B

(2001) Dimensional variability of nonstandardized greater

taper finger spreaders with matching gutta-percha points.


Brown WS, Dewey WA, Jacobs HR (1970) Thermal properties

of teeth. Journal of Dental Research 49, 752–5.

Burleson A, Nusstein J, Reader A, Beck M (2007) The in vivo

evaluation of hand/rotary/ultrasound instrumentation in

necrotic, human mandibular molars. Journal of Endodontics

33, 782–7.

Bystrom A, Sundqvist G (1981) Bacteriologic evaluation of the

efficacy of mechanical root canal instrumentation in end-

odontic therapy. Scandinavian Journal of Dental Research 89,

321–8.

Cameron JA (1987) The synergistic relationship between

ultrasound and sodium hypochlorite: a scanning electron

microscope evaluation. Journal of Endodontics 13, 541–5.

Cameron JA (1988) The effect of ultrasonic endodontics on the

temperature of the root canal wall. Journal of Endodontics 14,

554–9.

Cheung GS, Stock CJ (1993) In vitro cleaning ability of root

canal irrigants with and without endosonics. International


Cunningham WT, Martin H, Forrest WR (1982) Evaluation of

root canal debridement by the endosonic ultrasonic syner-

gistic system. Oral Surgery, Oral Medicine, and Oral Pathology

53, 401–4.

De Deus QD (1975) Frequency, location, and direction of the

lateral, secondary, and accessory canals. Journal of Endo-

dontics 1, 361–6.

Lumley PJ, Walmsley AD, Laird WR (1988) An investigation

into cavitational activity occurring in endosonic instrumen-

tation. Journal of Dentistry 16, 120–2.

Martin H (1976) Ultrasonic disinfection of the root canal. Oral

Surgery, Oral Medicine, and Oral Pathology 42, 92–9.

Martin H, Cunningham W (1985) Endosonics- the ultrasonic

synergistic system of endodontics. Endodontics and Dental

Traumatology 1, 201–6.

Naenni N, Thoma K, Zehnder M (2004) Soft tissue dissolution

capacity of currently used and potential endodontic irri-

gants. Journal of Endodontics 30, 785–7.

Nair PN, Henry S, Cano V, Vera J (2005) Microbial status of

apical root canal system of human mandibular first molars

with primary apical periodontitis after ‘‘one-visit’’ endodon-

tic treatment. Oral Surgery, Oral Medicine, Oral Pathology,

Oral Radiology, and Endodontics 99, 231–52.

Richman MJ (1957) The use of ultrasonics in root canal

therapy and resection. Journal of Dental Medicine 12, 12–8.

Sirtes G, Waltimo T, Schaetzle M, Zehnder M (2005) The

effects of temperature on sodium hypochlorite short-term

stability, pulp dissolution capacity, and antimicrobial effi-

cacy. Journal of Endodontics 31, 669–71.

van der Sluis LW, Wu MK, Wesselink PR (2005) The efficacy

of ultrasonic irrigation to remove artificially placed dentine

debris from human root canals prepared using instruments

of varying taper. International Endodontic Journal 38, 764–8.

van der Sluis LW, Versluis M, Wu MK, Wesselink PR (2007)

Passive ultrasonic irrigation of the root canal: a review of

the literature. International Endodontic Journal 40, 415–26.

Suslick KS (1990) Sonochemistry. Science 247, 1439–45.

Vianna ME, Horz HP, Gomes BP, Conrads G (2006) In vivo

evaluation of microbial reduction after chemo-mechanical

preparation of human root canals containing necrotic pulp

tissue. International Endodontic Journal 39, 484–92.

Walmsley AD (1987) Ultrasound and root canal treatment:

the need for scientific evaluation. International Endodontic

Journal 20, 105–11.



CLINICAL ARTICLE

A preliminary study of the use ofperipheral quantitative computedtomography for investigating root canalanatomy

M. T. Sberna1, G. Rizzo2, E. Zacchi3, P. Cappare1 & A. Rubinacci3

1Department of Odontoiatrics; 2IBFM-CNR, Department of Nuclear Medicine; and 3Bone

Metabolic Unit, Scientific Institute H San Raffaele, Milan, Italy

Abstract

Sberna MT, Rizzo G, Zacchi E, Cappare P, Rubinacci A. A preliminary study of the use of

peripheral quantitative computed tomography for investigating root canal anatomy. International Endo-

dontic Journal, 42, 66–75, 2009.

Aim To evaluate the use of peripheral quantitative computed tomography (pQCT) for

qualitative and quantitative analysis of root canal anatomy and for assessing the extent of

canal enlargement during root canal instrumentation.

Summary The volumevariation achievedbyS1ProTaper instruments in the coronal third of

the root canals was analysed using peripheral computed tomography. The tooth was

scanned in the horizontal plane producing 36 consecutive cross-sectional images. All images

were the result of 360 projections with a section thickness of 250 lm, a distance between

slices of 0.5 mm and an in-plane pixel size of 70 · 70 lm. The evaluation was completed

before and after S1 ProTaper instrumentation (with or without circumferential filing) of one

root canal of a freshly extracted maxillary first premolar tooth. The acquired images were

realigned geometrically and processed using a 3D visualization software. pQCT scanning

allowed 3D reconstruction of the root canal anatomy and the assessment of the extent of

canal enlargement during root canal instrumentationwith lateral displacement of canal walls

and hence volume change being greater than the coefficient of variation. The densitometry

evaluation showed uniform density along the root canal wall.

Key learning points

• pQCT scanning allowed 3D reconstruction of the root canal anatomy and the

assessment of the extent of canal enlargement during root canal instrumentation.

• pQCT shows promise for allowing qualitative and quantitative analysis of endodontic

procedures.

Keywords: 3D imaging, peripheral quantitative computed tomography, qualitative and

quantitative methodology, root canal instrumentation, root canal preparation.

Received 17 July 2007; accepted 2 June 2008

Correspondence: Alessandro Rubinacci, Bone Metabolic Unit, Scientific Institute H San

Raffaele, via Olgettina, 60, 20132 Milan, Italy (Tel.: +39 0226432320; fax: +39 0226433038;


International Endodontic Journal, 42, 66–75, 2009 ª 2008 International Endodontic Journal

doi:10.1111/j.1365-2591.2008.01452.x

66

Introduction

Precise morphological mapping of the root canal system is a prerequisite for the

evaluation of endodontic instruments and procedures. A detailed understanding of the

root canal system is, in fact, critical for the characterization of all factors that might have a

significant impact on the volume of root canals and pulp chambers and in the development

of successful therapeutic strategies. Conventional destructive approaches based upon 3D

computer-based reconstructions of histological sections do not allow a longitudinal

assessment of the endodontic therapy, are limited by the poor precision of the volumetric

algorithms, and do not permit a systematic mapping of the endodontic volumes (Walton

1976). Several non-destructive approaches have been developed more recently. For

example, computed tomography has been applied extensively to the detection of enamel

thickness from an anthropological perspective (Gantt et al. 2006), but the available

resolution did not allow a precise mapping of the root canal nor the estimation of the canal

volumes that were usually overestimated (Gantt et al. 2006). The inability of conventional

imaging techniques to visualize the root canal system drove the development of

alternative imaging modalities. In this context, two procedures have proven to be suitable

for the non-destructive exploration of both teeth and the volumetry of the root canals,

namely magnetic resonance microscopy (MRM) and X-ray computed microtomography

(lCT).

Magnetic resonance microscopy, a high-resolution magnetic resonance spectroscopy

system, constitutes a powerful tool for a detailed analysis of teeth without applying

ionizing radiation (Tseng et al. 2007). However, standard methodologies require a strong

proton signal of the surrounding liquid to produce a boundary surface image to visualize

the mineralized tissue. Magnetic resonance tomography (MRT) with stray field imaging

(STRAFI) (Baumann et al. 1993) can achieve this directly, in a very short T2 time, but with

poor resolution and the different hard tooth tissue components cannot be differentiated as

in the case of MRM. Imaging of all structural components of a tooth with one system and

one image was not possible until the demonstration that constant-time imaging (CTI)

techniques enabled the detection of magnetic signals from the hard tooth tissues, as well

as from the proton- and signal-intensive pulpal tissue. By presenting both signals in one

image with a resolution as low as 195 lm, CTI combines the advantages of both the

standard MRM and the STRAFI (Appel & Baumann 2002), but it might be limited in the

qualitative and quantitative description of the smallest components of the pulpal chamber.

The lCT is a miniaturized form of conventional computerized tomography. The lCT

scanner uses an X-ray tube as radiation source and a 3D reconstruction algorithm.

Recently, lCT has been introduced to evaluate not only cross-sections of roots but also 3D

shapes of canal systems at resolutions as high as 36 lm (Dowker et al. 1997, Bjørndal

et al. 1999, Rhodes et al.1999, Peters et al. 2000, 2001, Bergmans et al. 2001, Gluskin

et al. 2001, Gao et al. 2006, Lee et al. 2006). This innovation was achieved because new

hardware and software were available to evaluate the metrical data created by lCT, thus

allowing geometrical changes in prepared canals to be determined more precisely. This

technique has two disadvantages: it is limited to the processing of two extracted teeth at

a time because of the small size of the gantry and has a long scanning time up to 6 h

(Peters et al. 2003). New developments include high-resolution X-ray computed tomog-

raphy (HRXCT) and flat panel-based volume computed tomography (fpVCT). HRXCT is

applied to the 3D reconstruction of enamel thickness, and of dentine and pulp chamber

volumes at a resolution ranging from 5 to 100 lm by exporting two-dimensional digitized

images obtained by combining modular energy sources (125–450 kV) and modular

detectors (Gantt et al. 2006). The fpVCT has also been found suitable for the qualitative

visualization of the root canal system despite its low spatial resolution of 150 lm. This

CLIN

ICAL

ARTIC

LE


technique is in fact able to visualize dentine, enamel and the root canal system in 3D-

image reconstruction that, because of the size of the gantry (40 cm in diameter), might

include several teeth at the same scanning time (Hannig et al. 2006).

However, a systematic evaluation of endodontic instruments and procedures based

upon these instruments is not practicable, given their high cost and limited availability.

Therefore, the following study is designed to evaluate the feasibility of applying peripheral

quantitative computed tomography (pQCT) to the qualitative and quantitative analysis of

root canal anatomy and for assessing the extent of canal enlargement during root canal

instrumentation. pQCT has been originally designed for the diagnosis of osteoporosis in

humans, rats and mice (Schmidt et al. 2003). The unit works with a specially developed

X-ray tube having a minute focal spot whilst the detector system consists of a series of

miniature semiconductor crystals. The device is equipped with a special detector

collimator that can be switched up to four collimator sizes corresponding to the four

section thicknesses (100, 250, 500 and 750 lm). Although the planar resolution of pQCT

(70 · 70 lm) does not have the same resolution as lCT, it might provide a nondestructive

morphological investigation at low cost and shorter scanning times.


Specimen selection and preparation

The study is preliminary in nature, with only one tooth analysed. One root canal of a

maxillary first premolar tooth, freshly extracted for clinical reasons and not relating to this

study, was selected. After preparing a standard access cavity, the canal was passively

negotiated with sizes 10 and 20 K-files to the apical foramen; the working length was

determined visually.

Canal preparation was completed by a single operator using Ni–Ti rotary instrumenta-

tion; the S1 ProTaper (Dentsply Maillefer, Ballaigues, Switzerland) was mounted on an

ATR Tecnika Vision system (motor and handpiece) (ATR, Pistoia, Italy).

For assessing the extent of canal enlargement during root canal instrumentation, two

different applications of S1 ProTaper on the coronal third of the root canal were

considered: the first in compliance with the manufacturer’s protocol and the other

arbitrarily modified to produce a loss of the canal wall structure. Therefore, S1 ProTaper

was used, for approximately 9 s, until reaching 1 mm from the working length, centering

and avoiding any lateral movement (first phase), then applying lateral displacement

(second phase) for a further 9 s. The instrumentation was deliberately applied to one side

of the root canal system, leaving the other side untreated as a control.

Scanning

Tomographic tooth scanning and measurements were obtained before and after each

instrumentation phase.

A pQCT scanner was used for the measurements (Research SA+; Stratec Medizin-

technik GmbH, Pforzheim, Germany). This translation rotation scanner works with a

specially developed X-ray tube with a 50-lm spot size (high voltage 50 kV, anode current

<0.3 mA, mean X-ray energy 37 keV, energy distribution after filtration 18 keV full width

half maximum [FWHM]). The detector-system consists of 12 miniature semiconductor

crystals with amplifiers. The precision error supplied by the manufacture for density

measurement in vivo is around 1.5%.

The tooth was scanned in the horizontal plane producing 36 consecutive cross-sectional

images at a distance of 0.5 mm.

CLIN

ICAL

ARTIC

LE


All images were obtained with 360 projections, with a section thickness of 250 lm and

at an in-plane pixel size of 70 · 70 lm at a scan speed of 3 mm s)1. Total scanning time

was 4 h. Operator time was limited to 10 min.

Scanning of a selected region of the root canal was also obtainedwith a section thickness

of 100 lm.As no additional informationwas available, this scanning procedurewas aborted.

To orientate the long axes of the tooth parallel to the image planes, the tooth was fixed

with manufacturer-made plastic holders. The correct longitudinal positioning was

determined by means of an initial scout scan.

Qualitative data analysis

To correct the possible mispositioning of the specimen in the pQCT gantry, the acquired

studies were geometrically realigned using a registration technique based on the

maximization of mutual information implemented in a home-made software package

(Rizzo et al. 2005). After registration, the studies corresponded geometrically, with sub-

voxel accuracy, and could be compared correctly.

The registered images were then processed using 3D visualization software (amira 4.1;

Mercury Computer System Inc., Chelmsford, MA, USA) to generate 3D rendering of the

tooth external surface and the root canal, for the qualitative evaluation of the modification

of the root canal size, produced by the S1 ProTaper. The coefficient of variation of the

reconstructed volumes after repositioning was assessed by performing the quantitative

analysis on the actual cross-sections of the root canal that had not undergone to

instrumentation. It varied from 3.3% to 7.1%.

Quantitative data analysis

A quantitative assessment of the canal volume variations induced by the instrumentation

was carried out using the 3D analysis software ‘analyze’ (Biodynamic Research Unit,

Mayo Clinic, Rochester, MN, USA) (Robb et al. 1989). From each registered image, the

root canal volume, corresponding to the area of interaction of S1 ProTaper (coronal third of

the root canal), was extracted, by calculating, on each slice, the isocontour corresponding

to the same isovalue. The volume of the dentine removed was then obtained by

subtracting the canal volumes after and before treatments.

Furthermore, a densitometry evaluation was performed, with the pQCT scanner, which

directly provided sectional images accurately calibrated in terms of density. Dentine

density in each scan was calculated by analyze. For this purpose, densities <500 mg cm)3

corresponding to the tooth canal or >2000 mg cm)3 corresponding to the tooth enamel

were excluded.

Results

Qualitative evaluation

In Fig. 1, the 3D representation of the external tooth surface and the root canal is shown

for each study, after spatial registration. From the qualitative analysis it is possible to note

the effect of using ProTaper in a lateral displacement mode: the size of the treated canal

area is enlarged (see arrow).

The same effect can be seen in Fig. 2, which visualizes the surface of the canals, before

and after the treatments. The increment of canal volume is evidentwhenProTaper is used in

a lateral displacement mode (Fig. 2e) but is not noticeable when ProTaper is used more

passively (Fig. 2d).

CLIN

ICAL

ARTIC

LE


Figure 1 Representative 3D rendering of the external tooth surface and canal for each experimental

condition. Left: treatment with K-file 20 instrumentation. Middle: treatment with S.1 ProTaper

instrumentation used as protocol. Right: treatment with S.1 ProTaper instrumentation used by lateral

displacement. The enlargement of treated area (arrow) is clearly visible in the third situation.

(a) (b)

(d) (e)

(c)

Figure 2 Comparison of 3D rendering of the canal surfaces in different conditions. Top row: (a) K-file

20 (white), (b) S1 ProTaper as protocol (green), (c) S1 ProTaper with lateral displacement (red). Bottom

row: (d) superposition of S1 ProTaper as protocol and K-file 20, (e) superposition of S1 ProTaper as

lateral displacement and K-file 20. In this last figure, the enlargement of canal using S1 ProTaper as

lateral displacement results in an evident red area on the reference K-file 20 white surface.

CLIN

ICAL

ARTIC

LE


Quantitative evaluation

Figure 3 shows orthogonal views corresponding to the 3D conditions after registration,

with the region of interest (treated area) superimposed as red overlay.

The measured volume sizes for each slice belonging to the treated area are shown in

Fig. 4. Lateral displacement produced volume changes far above those detected when

ProTaper is used more passively. By applying lateral displacement between the K-file and

S1 used in a brushing mode, volume changes ranged from +0.09 mm3 (38 voxels) in slice

Figure 3 Tooth cross-sectional views corresponding to the three different experimental conditions.

Top row: treatment with K-file 20. Middle row: treatment with S1 ProTaper used as protocol. Bottom

row: treatment with S1 ProTaper used as lateral displacement. The red overlay corresponds to the

treated area.

CLIN

ICAL

ARTIC

LE


23 to +0.18 mm3 (72 voxels) in slice 26, whereas without lateral displacement volume

changes ranged from )0.030 mm3 (or 12 voxels) in slice 23 to +0.049 mm3 (or 20 voxels)

in slice 26.

The measured volume sizes for the reconstructed treated area were 2.56 mm3 before

S1 ProTaper, 2.63 mm3 after S1 ProTaper without lateral displacement, and 3.10 mm3

after S1 ProTaper with lateral displacement. This resulted in an increment of 2.66% of the

canal volume when ProTaper was used more passively, and in an increment of 21% when

lateral displacement was applied. Whilst the former increment was in the range of the

coefficient of variation, as assessed on the root canal not undergoing to instrumentation

(3.3–7.1%), the latter was far above. As not more than one tooth was used, no significant

statistical calculation can be presented.

The cross sectional densitometric analysis was able to identify the dentinoenamel

junction that was represented in Fig. 5. The density distribution pathway clearly

distinguished dentine density from enamel. The longitudinal densitometric analysis

globally shows no differences in the density of dentine: as illustrated in Fig. 6, the profile is

flat and only small density gradients can be observed for the first curve points, principally

because of partial volume effects induced by slice thickness.

Figure 4 Comparison of tooth canal volumes in the three different experimental conditions. Volumes

were measured in all the slices belonging to the treated area (slice 23–26, total longitudinal extension:

2000 lm).

Figure 5 Typical cross-sectional density profile showing the density of the enamel and dentine. The

abrupt drop in the density profile corresponds to the dentinoenamel junction.

CLIN

ICAL

ARTIC

LE


Discussion

The study has shown that pQCT instruments, developed for bone mineral analysis and

having a spatial resolution of 70 · 70 · 250 lm, show promise for allowing a precise and

reliable mapping of the root canal system by producing contiguous slices of teeth. The 3D-

image reconstruction and the measurements of volumes and densities obtained by pQCT

scanning appear to be suitable for the qualitative and quantitative assessment of the

changes in root canal shape following instrumentation. The application of a pQCT system to

endodontic imaging offers advantages over current NMR and lCT techniques, mainly

relating to its lower cost and wider availability, whilst scanning time is only marginally

reduced (4 h vs. 6 h). pQCT allows direct visualization of tooth tissues, i.e. dentine, enamel

and root canal system,which are clearly distinguishable in the 3D images, and candetermine

the impact of spatial distribution of the dental volumetric density (enamel versus dentine) on

dental pathology. This has not been systematically evaluated and offers potential

advantages on our current understanding of the genetic and environmentally related

differences in dentine signalling that could alter enamel structure with an impact on dental

health. A minor advantage of pQCT is related to its large scanner gantry opening (9 cm) that

can include several teeth at the same scanning time as fpVCT (Hannig et al. 2006), whereas

lCT systems can allowevaluation of two teeth only (Peters et al. 2001). The acquired data at

each repositioning were realigned geometrically before comparison using a registration

technique based on the maximization of mutual information implemented in a home-made

software package (Rizzo et al. 2005). As the registered sections were geometrically aligned

with sub-voxel accuracy, the comparisons should be considered precise and reliable. The

detection of volume changes at the side of the root canal system, only where they were

expected to be found after adequate instrumentation, sustains a potential application of the

developed methodology in the clinical setting. When pQCT, lCT and histomorphometry

were compared, the results showed that a pQCT scanning at 500-lm thickness can yield

satisfactory precision and accuracy in microstructural representation of the scanned bone

site (Schmidt et al. 2003). The highest agreement was found between pQCT and lCT being

based on the measurement of the same physical property as X-ray absorption.

However, a rigorous analysis of the limits of the pQCT in reconstructing small canals

should be outlined. The larger pixel size (70 · 70 lm), and the consequent lower

resolution of the pQCT versus lCT, introduces larger partial volume effects that might

Figure 6 Dentine density measured in the scan correspond to K-file 20 treatment. Slices 1 and 36 are

not considered, as they contain very few voxels of dentine.

CLIN

ICAL

ARTIC

LE


affect the volumetric evaluation of the pulp chamber and root canals as well as the

definition of the dentinoenamel junction and the derived parameters as density. Even if

the object of theoretical correction (Rittweger et al. 2004), this limitation is critical and is

applied when projecting a continuous object on a discrete grid. Hence, quantitative

image analysis is prone to errors where the edge of the object is within the sampling

grid. The segmentation process can overestimate sharp edges depending upon the voxel

volumes as the partial volume effect increases with voxel size (i.e. at lower image

resolution) and decreases with the object size. This implies that side and/or accessory

canals as well as main canals smaller than the voxel size are not detectable with

sufficient accuracy, thus hampering correct visualization and analysis. The spatial

resolution of the pQCT methodology applied is critical to the reliability of volume change

measurements. It is generally accepted that half voxel size for each voxel forming the

volume surface contributes to the uncertainty range in volume estimate, because of

spatial resolution. It follows that, in the case of the root canal system, the uncertainty

varies along the canal and is related to its size. By approximating to a cylinder the shape

of the root canal represented in a single tomographic slice, the measurement uncertainty

of the volume changes observed should be expected ranging from 10% to 11% (from 28

to 31 voxels) for canal volume ranging from 250 to 298 voxels (slices 23–26). This

quantitative analysis suggests that the volume changes observed when lateral displace-

ment was applied were reliable and, as a consequence, lateral displacement should be

avoided to preserve optimal mechanical strength of tooth. In fact, the potential for the

root filled teeth to fracture increases proportionally with the amount of dentine removed

(Pilo et al. 1998).

Conclusion

This study has presented an innovative and nondestructive methodology to illustrate canal

morphology and canal volume changes after instrumentation by applying pQCT analysis.

The 3D reconstruction methodology based on pQCT images described here deserves

further systematic evaluation to fully validate its application in the clinical setting as a tool

for qualitative and quantitative analysis of the endodontic procedures.

Disclaimer

Whilst this article has been subjected to Editorial review, the opinions expressed, unless

specifically indicated, are those of the author. The views expressed do not necessarily

represent best practice, or the views of the IEJ Editorial Board, or of its affiliated Specialist

Societies.

Acknowledgements

Wewish to acknowledge Johannes Willnecker, Stratec Medizintechnik GmbH, Pforzheim,

Germany, for providing the technical information required for the development of the study.

References

Appel TR, Baumann MA (2002) Solid-state nuclear magnetic resonance microscopy demonstrating

human dental anatomy. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and


Baumann MA, Doll GM, Zick K (1993) Stray-field imaging (STAFI) of teeth. Oral Surgery, Oral

Medicine, and Oral Pathology 75, 517–22.

CLIN

ICAL

ARTIC

LE


Bergmans L, Van Cleynenbreugel J, Wevers M, Lambrechts P (2001) A methodology for quantitative

evaluation of root canal instrumentation using micro computed tomography. International


Bjørndal L, Carlsen O, Thuesen G, Darvann T, Kreiborg S (1999) External and internal macromorphol-

ogy in 3D-reconstructed maxillary molars using computerized X-ray microtomography. International


Dowker S, Davis G, Elliott J (1997) X-ray microtomography-non-destructive three-dimensional imaging

for in vitro endodontic studies. Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology, and


Gantt DG, Kappleman J, Ketcham RA, Alder ME, Deahl TH (2006) Three-dimensional reconstruction of

enamel thickness and volume in humans and hominoids. European Journal of Oral Sciences 114,

360–4.

Gao Y, Fan B, Cheung GS, Gutmann JL, Fan M (2006) C-shaped canal system in mandibular second

molars part IV: 3-D morphological analysis and transverse measurement. Journal of Endodontics 32,

1062–5.

Gluskin AH, Brown DC, Buchanan LS (2001) A reconstructed computerized tomography comparison

of Ni–Ti rotary GT files versus traditional instruments in canals shaped by novice operators.


Hannig C, Krieger E, Dullin C et al. (2006) Volumetry of human molars with flat panel-based volume CT

in vitro. Clinical Oral Investigations 10, 253–7.

Lee JK, Ha BH, Choi JH, Heo SM, Perinpanayagam H (2006) Quantitative three-dimensional analysis

of root canal curvature in maxillary first molars using micro-computed tomography. Journal of


Peters OA, Laib A, Ruegsegger P, Barbakow F (2000) Three dimensional analysis of root canal

geometry using high resolution computed tomography. Journal of Dental Research 79, 1405–9.

Peters OA, Schonenberger K, Laib A (2001) Effects of four Ni–Ti preparation techniques on root canal

geometry assessed by micro computed tomography. International Endodontic Journal 34, 221–30.

Peters OA, Peters CI, Shonenberger K, Barbakow F (2003) ProTaper rotary root canal preparation:

effects of canal anatomy on final shape analysed by micro CT. International Endodontic Journal 36,

86–92.

Pilo R, Corcino G, Tamse A (1998) Residual dentin thickness in mandibular premolars prepared with

hand and rotary instruments. Journal of Endodontics 24, 401–4.

Rhodes JS, Pitt Ford TR, Lynch PJ, Liepins PJ, Curtis RV (1999) Micro-computed tomograpy: a new

tool for experimental endodontology. International Endodontic Journal 32, 165–70.

Rittweger J, Michaelis I, Giehl M, Wusecke P, Felsenber D (2004) Adjusting for the partial volume

effect in cortical bone analyses of pQCT images. Journal of Musculoskeletal and Neuronal

Interactions 4, 436–41.

Rizzo G, Castiglioni I, Arienti R et al. (2005) Automatic registration of PET and CT studies for clinical

use in thoracic and abdominal conformal radiotherapy. Quarterly Journal of Nuclear Medicine 49,

267–79.

Robb RA, Hanson DP, Karwoski RA, Larson AG, Workman EL, Stacy MC (1989) Analyze: a

comprehensive operator-interactive software package for multi-dimensional medical image display

and analysis. Computerized Medical Imaging and Graphics 13, 433–54.

Schmidt C, Priemel M, Kohler T et al. (2003) Precision and accuracy of peripheral quantitative

computed tomography (pQCT) in the mouse skeleton compared with histology and microcomputed

tomography (CT). Journal of Bone and Mineral Research 18, 1486–96.

Tseng YH, Tsai YL, Tsai TW et al. (2007) Double-quantum filtered heteronuclear correlation

spectroscopy under magic angle spinning. Solid State Nuclear Magnetic Resonsance 31, 55–61.

Walton RE (1976) Histologic evaluation of different methods of enlarging the pulp canal space. Journal


CLIN

ICAL

ARTIC

LE


CASE REPORT

Endodontic management of badlybroken down teeth using the canalprojection system: two case reports

A. S. Bhomavat, R. K. Manjunatha, R. N. Rao & K. H. KidiyoorDepartment of Conservative Dentistry and Endodontics, S.D.M. College of Dental Sciences and

Hospital, Dharwad, Karnataka, India

Abstract

Bhomavat AS, Manjunatha RK, Rao RN, Kidiyoor KH. Endodontic management of

badly broken down teeth using the canal projection system: two case reports. International Endodontic

Journal, 42, 76–83, 2009.

Aim Teeth that have been weakened by caries and require root canal treatment to

maintain their functional integrity may present with minimal coronal tooth structure and

are a challenge for isolation and restoration. The aim of this clinical report is to

demonstrate the management of badly broken down teeth using the Projector Endodontic

Instrument Guidance System (PEIGS).

Summary The PEIGS is an adjunct to root canal treatment designed to enhance the ease

of treatment delivery. Use of this system facilitates projection of canal orifices from the

floor of the pulp chamber to the cavosurface, providing direct visualization of and physical

access to the projected canals. This report demonstrates the use of this novel device for

the management of two badly broken down teeth.

Key learning points

Use of the endodontic projection system has the following advantages:

• ‘Projects’ the canal orifice from the floor of the pulp chamber to the cavosurface,

thereby enhancing visualization and access to the canals.

• The bonded coronal build up reduces the risk of interappointment crack initiation and

coronal-radicular fracture of weakened tooth structure.

• Permits individualization of canals especially when they lie in close proximity to each

other on the chamber floor.

• Isolation may be facilitated by ease of clamp retention, rendering many structurally

debilitated teeth endodontically treatable.

Keywords: broken down teeth, endodontic canal projection, isolation, pre-endodontic

build-up, projector.

Received 15 September 2007; accepted 13 July 2008

doi:10.1111/j.1365-2591.2008.01465.x

Correspondence: Dr Anisha S. Bhomavat, Department of Conservative Dentistry and

Endodontics, S.D.M. College of Dental Sciences and Hospital, Sattur, Dharwad – 580009,

Karnataka, India (Tel.: +91 9314130001, +91 22 28330846; fax: +91 836 2467612;



Introduction

Technical and scientific advances in endodontics have resulted in retention of teeth, which

were earlier deemed untreatable (Johns et al. 2006). It is universally accepted that

preservation of a natural tooth with a good prognosis is superior to tooth loss and

replacement (Roda & Gettleman 2006).

The current techniques employed tomanage severely broken down teeth include the use

of special clamps with specific designs, surgical exposure of the cervical tooth structure to

facilitate clamp placement, use of orthodontic bands, preformed copper bands, pin or

adhesive retained amalgam, composite and glass ionomer buildups. However, these have

inherent disadvantages (Madisonet al.1986, Jeffrey&Woolford1989). Presenceofminimal

coronal structure can risk further damage to the crown during rubber dam clamp placement

thereby compromising isolation and causing subsequent coronal leakage (Jeffrey &

Woolford 1989, Zerr et al. 1996). Pre-endodontic build-up of the coronal tooth structure

following caries removal and identification of the canal orifices can facilitate the endodontic

process by providing a strong core and coronal seal (Kurtzman 2004).

The canal projection technique using the Projector Endodontic Instrument Guidance

System (PEIGS) (CJM Engineering, Santa Barbara, CA, USA) provides pre-endodontic

reconstruction of debilitated coronal and radicular tooth structure whilst preserving

individualized access to canals (Kurtzman 2004, http://www.cjmengineering.com). This

case report introduces the innovative concept of using the ‘Projector’ which ‘projects’ the

canal orifices from the chamber floor to the cavosurface providing better visibility and

access (Weathers 2004), and also ensures optimum isolation and reinforcement of the

tooth structure.

Case reports

Case report 1

A 36-year-old female reported to the Department of Conservative Dentistry and

Endodontics, S.D.M. College of Dental Sciences, Dharwad, India, complaining of a dull,

mild intermittent pain in the right maxillary posterior region for 2 months. Intra-oral

examination revealed the presence of a grossly decayed tooth, 16 (FDI), with three walls

missing (Fig. 1a). Pulp sensibility testing elicited a negative response. The preoperative

radiograph (Fig. 1b) revealed deep occlusal caries involving the pulp and widening of the

periodontal ligament space in relation to the palatal root. A diagnosis of pulpal necrosis and

chronic periradicular periodontitis was made.

Root canal treatment was then planned using the PEIGS as rubber dam isolation was

challenging. The Projector is a small, black, cone-shaped plastic device, which slides onto

an endodontic file (Fig. 1d). It has a central lumen, an apical bevel and is made of a

specially formulated plastic (linear low-density polyethylene) which is nonadherent to

dental restorative materials. It is available in two sizes; ‘regular’ which is used in cases

where the size of the access cavity is adequate to accommodate the medium-sized

device, and ‘skinny’ which is used in cases where the size of the access cavity is not

adequate to accommodate the medium-sized device (Table 1).

After securing adequate anaesthesia and application of rubber dam with a clamp with

apically inclined beaks, caries was excavated. Access cavity preparation was performed

and four canal orifices were identified (Fig. 1c). The canals were enlarged to a size 20 file

using the standardized method of cleaning and shaping. Canal orifices were dimpled with

a slow speed round bur (Mani Inc., Tochigi-Ken, Japan) of diameter 1 mm, to facilitate

placement of the projectors and to prevent flow of adhesive into the canals.

CASE

REPORT


A stainless steel automatrix band (Hawe Supermat�; KerrHawe, Lugano, Switzerland)

was placed followed by the application of phosphoric acid gel (Scotchbond Etchant gel;

3M ESPE, St Paul, MN, USA) to etch the exposed dentine and enamel. Rinsing and drying

was accomplished after 30 s. The Projectors were placed on four endodontic files and slid

up toward the file handles, so that 5–8 mm of each file tip protruded beyond the tip of the

Projector. Different sizes of files were used to aid in identification of the projected orifices.

Size 20 was used for the mesiobuccal canal, size 15 for the second mesiobuccal canal,

size 25 for the distobuccal canal and size 30 for the palatal canal. Each file with a Projector

was then inserted into its respective orifice and the Projector was pressed into place with

cotton pliers until it seated precisely and snugly into the dimple created at the orifice. A

dentine bonding agent (Adper Single Bond, 3M ESPE) was then applied and light-cured.

(a)

(g) (h) (i)

(b)

(d) (e) (f)

(c)

Figure 1 (a) Preoperative photograph: severely broken down tooth 16 (mirror view). (b) Preoperative

radiograph: deep occlusal caries and chronic periradicular periodontitis, tooth 16. (c) Access opening

completed under rubber dam, four orifices detected. (d) Files are prepared with projectors. (e)

Composite built up around projectors to occlusal surface. (f) Files removed leaving projectors in place.

(g) Projectors are removed using H-file. (h) Final result: orifices projected to occlusal surface. (i)

Postobturation radiograph.

Table 1 Details of the dimensions of the PEIGS

Regular

Overall length c. 10.00 mm

Diameter 1 mm from apical end c. 1.20 mm

Large diameter c. 2.00 mm

Tapered lumen full length

Skinny

Overall length c. 13.00 mm

Diameter 1 mm from apical end c. 0.80 mm

Large diameter c. 1.14 mm

Tapered lumen full length

PEIGS, Projector Endodontic Instrument Guidance System.

CASE

REPORT


The build-up was placed in increments using a hybrid composite (Filtek Z100, 3M ESPE)

and light-cured (Fig. 1e).

Following curing, the files were removed by counter-rotation, leaving the Projectors in

place (Fig. 1f). A high speed, bull-nosed diamond (Mani, Inc) was used to level the occlusal

surface providing ideal endodontic reference points. The final result was a stable coronal

structure with straight-line access into each canal with maximum structural reinforce-

ment. A size 60 Hedstrom hand file was then used to remove the Projectors from the

core, by rotating it clockwise, to engage the flutes in the lumen of each Projector and

withdrawing (Fig. 1g). Thus, a pre-endodontic build-up with individualized access to each

canal was achieved successfully (Fig. 1h).

The original hand file was introduced into each projected orifice and a working length

radiograph was taken. Standard instrumentation was performed to clean and shape the

canals. Interim coronal seal of the canals was simplified by snipping 3 mm from the large

diameter end of each Projector, reinserting them into their respective projected orifices

and then sealing each with Cavit (3M ESPE). At the subsequent visit, the small Cavit seals

were removed with a round bur, and the submerged Projectors were easily removed by

engaging them with a Hedstrom file and withdrawing. Following canal preparation and

filling to the level of the chamber floor (Fig. 1i), the composite in the projected canals was

freshened with a diamond bur (Mani, Inc.) and additional composite resin was bonded

directly over gutta percha to the level of the cavosurface. The pre-endodontic build-up

itself was used as a core and full crown preparation was performed followed by crown

cementation at a subsequent appointment.

Application of this technique created a conical projected orifice which was easily

visualized and accessed and consistently delivered the tip of the endodontic file to the

respective canal whilst maintaining independence of canals from each other. This

technique, once mastered, takes minimal time and greatly enhances treatment of badly

broken down teeth.

Case report 2

A 21-year-old female attended with the complaint of a mildly painful tooth in the

mandibular right posterior region for the past 4 months. Intra-oral examination revealed a

grossly decayed tooth, 46 (FDI). Pulp sensibility tests elicited a negative response. The

preoperative radiograph showed deep occlusal caries involving the pulp space and slight

widening of the periodontal ligament space. The pulp was diagnosed as necrotic,

associated with chronic periradicular periodontitis. Root canal treatment was initiated

using the PEIGS. The procedure for management of this badly broken down tooth was

similar to that described above. Figure 2a–e demonstrates the steps undertaken.

Discussion

The dentist may often be confronted with severely compromised teeth. High quality root

canal treatment and reconstructive procedures are prerequisites to ensure long-term

maintenance of such teeth (Ricucci & Grosso 2006). In such difficult cases, canal

Projectors can facilitate adequate access and preparation of root canals during root canal

treatment. This technique enhances management of complexities including severe

coronal breakdown, tipped/rotated teeth, limited mouth opening and near proximity of

orifices on the chamber floor (Weathers 2004).

In cases of severe coronal breakdown, various methods of isolation have been

suggested, including the use of clamps with apically inclined beaks, the Silker-Glickman

clamp (The Smile Center, Deerwood, MN, USA), or the split-dam technique (Kurtzman

CASE

REPORT


2004). However, multiple tooth isolation can be less effective than single tooth isolation

and often requires the use of other aids such as floss ligation and/or sealants (Scott 2002).

Occasionally, periodontal or restorative procedures may be necessary to simplify

placement of the rubber dam (Ingle et al. 2002). These procedures include clamping of

anaesthetized attached gingiva, surgical crown lengthening procedure such as gingivopl-

asty or alveoloplasty (Gutmann & Lovdahl 1997) and the composite ‘donut’ technique

(Heydrich 2005). Restorative methods may also be considered to build up the tooth so that

a retainer can be placed properly (Lovdahl & Gutmann 1980, Lovdahl & Wade 1997). A

preformed copper or orthodontic band or a temporary crown may be cemented over the

remaining natural crown. However, the disadvantages include inferior sealing ability,

blockage of canal systems by cement during access opening or instrumentation and

periodontal inflammation if improperly placed/contoured.

(a)

(c) (d)

(b)

(e)

Figure 2 (a): Preoperative photograph: severely broken down tooth 46. (b) Preoperative radiograph:

deep occlusal caries and periradicular periodontitis. (c) Access opening completed under rubber dam,

three orifices detected, matrix band placed. (d) Final result: orifices projected to occlusal surface. (e)

Postobturation radiograph.

CASE

REPORT


Occasionally, so little tooth structure remains that even band or crown placement is

not possible. In such cases, it becomes necessary to replace missing tooth structure to

facilitate placement of the rubber dam clamp to prevent contamination of the working

field (Lovdahl & Gutmann 1980, Lovdahl & Wade 1997, Scott 2002). The tooth can be

built up with hard, fast-setting temporary cement (e.g. Ketac-Fil, ESPE, Seefeld,

Germany; TERM, LD Caulk, Milford, DE, USA), pin-retained amalgam or composites

(Ingle et al. 2002, Scott 2002). However, these restorative methods are time consuming;

they can impede endodontic access and may require replacement when they are

weakened by endodontic access procedures.

To overcome these challenges, the canal projection technique was developed and

offers the following advantages: (i) it ‘projects’ the canal orifice from the floor of the pulp

chamber to the cavosurface, thereby enhancing visualization and access to the canals, (ii)

permits individualization of canals and therefore can simplify management of canals that

lie in close proximity to each other on the chamber floor, (iii) can allow for ease of

isolation as canal projection essentially replaces missing tooth structure thereby

facilitating clamp retention and thus rendering many structurally debilitated teeth

treatable and (iv) allows files to be inserted easily, particularly nickel–titanium files which

are sometimes difficult to insert into mesial canals as they are unable to retain a bend, as

the canals are no longer obscured by prominent marginal ridges and other visual

obstructions.

The bonded composite coronal build-up decreases coronal leakage (Uranga et al. 1999,

Heling et al. 2002, Schwartz & Fransman 2005) and also reduces the risk of coronal-

radicular fracture during endodontic therapy thereby reinforcing the tooth (Hurmuzlu et al.

2003, Daneshkazemi 2004). Furthermore, the bonded core seals the accessory canals that

exit the chamber floor (Niemann et al.1993, Luglie & Sergente 2001, Haznedaroglu et al.

2003), providing a degree of protection to the chamber floor in cases where extensive

decay has left an area of the floor thin. This prevents leakage of contaminants to the

furcation through what would otherwise be a temporary seal between treatment visits.

The technique can also reinforce perforation repairs by overlaying mineral trioxide

aggregate (MTA) with a bonded resin prior to root canal treatment, preventing

re-aggravation of the perforation site during subsequent procedures (Ford et al. 1995).

Canal projection allows correction of misdirected access cavities by essentially

reconstructing the walls and floors around Projectors which act as ‘internal matrix

barriers’. It insulates files from metallic coronal restorations to facilitate accurate electronic

length determination (Carrotte 2004, Kim & Lee 2004) and also prevents ingrowth of

tissues in cases where cervical tooth structure has been destroyed. The canal projection

process elongates the ‘hydraulic chamber’ of each canal, offering advantages during the

hydraulic condensation of obturating materials, especially whilst using warm vertical

condensation techniques (Glickman & Pettiette 2006).

It should be noted that, as with many useful techniques, canal projection is a technique-

sensitive procedure and may have its limitations; in fact, the obturation may not be limited

to the canal orifices and initially it may be time consuming. However, once mastered, the

technique can be performed with speed and precision, and it can significantly enhance the

balance of treatment, particularly in cases of severe coronal break down.

Conclusion

Management of teeth with minimal coronal structure can be a challenging task when

root canal treatment is required as a part of oral rehabilitation. Coronal leakage, isolation

complexities and risk of interappointment coronal-radicular fracture may be major

contributors to endodontic failure. This case report demonstrates the use of an

CASE

REPORT


innovative technique, canal projection, as an efficient method for managing these

complex cases.

Conflict of interest

The authors affirm that they have no commercial interest in the materials used or their

method of use as discussed in this manuscript.

Acknowledgements

The authors would like to thank Dr C. John Munce for providing the Projector Endodontic

Instrument Guidance System for undertaking this case report as well as his keen guidance

in the preparation of this manuscript. Thanks to Dr Bhasker Rao, Principal, S.D.M. College

of Dental Sciences, Dharwad, India for his kind cooperation and support.

Disclaimer




Societies.

References

Carrotte P (2004) Endodontics: Part 7. Preparing the root canal. British Dental Journal 197, 603–13.

Daneshkazemi A R (2004) Resistance of bonded composite restorations to fracture of endodontically

treated teeth. The Journal of Contemporary Dental Practice 5, 51–8.

Glickman GN, Pettiette MT (2006) Preparation for treatment. In: Cohen S, Hargreaves K, eds.

Pathways of the Pulp, 9th edn. St Louis, MO, USA: Mosby, pp. 97–135.

Gutmann JL, Lovdahl PE (1997) Problems encountered in tooth isolation and access to the pulp

chamber space. In: Gutmann JL, Dumsha TC, Lovdahl PE, Hovland EJ, eds. Problem-Solving in

Endodontics: Prevention, Identification and Management, 3rd edn. St Louis, MO, USA: Mosby, pp.

47–67.

Haznedaroglu F, Ersev H, Odabasi H et al. (2003) Incidence of patent furcal accessory canals in

permanent molars of a Turkish population. International Endodontic Journal 36, 515–9.

Heling I, Gorfil C, Slutzky H, Kopolovic K, Zalkind M, Slutzky-Goldberg I (2002) Endodontic failure

caused by inadequate restorative procedures: review and treatment recommendations. The Journal

of Prosthetic Dentistry 87, 674–8.

Heydrich R W (2005) Pre-endodontic treatment restorations. A modification of the ‘donut’ technique.

Journal of the American Dental Association 136, 641–2.

Hurmuzlu F, Kiremitci A, Serper A, Altundasar E, Siso SH (2003) Fracture resistance of endodontically

treated premolars restored with ormocer and packable composite. Journal of Endodontics 29, 838–

40.

Ingle JI, Walton RE, Malamed SF, et al. (2002) Preparation for endodontic treatment. In: Ingle JL,

Bakland LK, eds. Endodontics, 5th edn. Hamilton, Ontario, Canada: B. C. Decker, pp. 357–404.

Jeffrey IW, Woolford MJ (1989) An investigation of possible iatrogenic damage caused by metal

rubber dam clamps. International Endodontic Journal 22, 85–91.

Johns BA, Brown LJ, Nash KD, Warren M (2006) The Endodontic Workforce. Journal of Endodontics

32, 838–46.

Kim E, Lee SJ (2004) Electronic apex locator. Dental Clinics of North America 48, 35–54.

Kurtzman GM (2004) Restoring teeth with severe coronal breakdown as a prelude to endodontic

therapy. Endodontic Therapy 4, 21–2.

CASE

REPORT


Lovdahl PE, Gutmann JL (1980) Periodontal and restorative considerations prior to endodontic therapy.

General Dentistry 28, 38–45.

Lovdahl PE, Wade CK (1997) Problems encountered in tooth isolation and periodontal support for the

endodontically compromised tooth. In: Gutmann JL, Dumsha TC, Lovdahl PE, Hovland EJ, eds.

Problem-Solving in Endodontics: Prevention, Identification and Management, 3rd edn. St Louis, MO,

USA: Mosby, pp. 203–27.

Luglie PF, Sergente C (2001) SEM study of morphology and incidence of accessory canals in the

furcation region of permanent molars. Minerva Stomatologica 50, 63–9.

Madison S, Jordan RD, Krell KV (1986) The effects of rubber dam retainers on porcelain fused to-metal

restorations. Journal of Endodontics 12, 183–6.

Niemann RW, Dickinson GL, Jackson CR, Wearden S, Skidmore AE (1993) Dye ingress in molars:

furcation to chamber floor. Journal of Endodontics 19, 293–6.

Pitt Ford TR, Torabinejad M, McKendry DJ, Hong CU, Kariyawasam SP (1995) Use of mineral trioxide

aggregate for repair of furcal perforations. Oral Surgery, Oral Medicine, Oral Pathology, Oral


Ricucci D, Grosso A (2006) The compromised tooth: conservative treatment or extraction? Endodontic

Topics 13, 108–22.

Roda RS, Gettleman BH (2006) Non surgical retreatment. In: Cohen S, Hargreaves K, eds. Pathways

of the Pulp, 9th edn. St Louis, MO, USA: Mosby, pp. 944–1010.

Schwartz RS, Fransman R (2005) Adhesive dentistry and endodontics: materials, clinical strategies

and procedures for restoration of access cavities: a review. Journal of Endodontics 31, 151–65.

Scott GL (2002) Isolation. In: Walton RE, Torabinejad M, eds. Principles and Practice of Endodontics,

3rd edn. Philadelphia, PA, USA: W.B. Saunders, pp. 118–29.

Uranga A, Blum JY, Esber S, Parahy E, Prado C (1999) A comparative study of four coronal obturation

materials in endodontic treatment. Journal of Endodontics 25, 178–80.

Weathers A K (2004) Endodontics from access to success, part 1 access, the important first step.

Dentistry Today 23, 78–85.

Zerr M, Johnson WT, Walton RE (1996) Effect of rubber-dam retainers on porcelain fused to metal.

General Dentistry 44, 132–4.

CASE

REPORT


CASE REPORT

Pulp revascularization of necroticbilateral bicuspids using a modifiednovel technique to eliminate potentialcoronal discolouration: a case report

K. Reynolds, J. D. Johnson & N. CohencaDepartment of Endodontics, School of Dentistry, University of Washington, Seattle, WA,

USA

Abstract

Reynolds K, Johnson JD, Cohenca N. Pulp revascularization of necrotic bilateral bicuspids using

a modified novel technique to eliminate potential coronal discolouration: a case report. International

Endodontic Journal, 42, 84–92, 2009.

Aim To present a case report in which the pulp of two bilateral mandibular premolars with

dens evaginatus were revascularized using a modified novel technique to avoid undesired

crown discolouration.

Summary Recently, regeneration of necrotic pulps has become an alternative conser-

vative treatment option for young permanent teeth with immature roots and is a subject of

great interest in the field of endodontics. This novel procedure exploits the full potential of

the pulp for dentine deposition and produces a stronger mature root that is better able to

withstand the forces than can result in fracture. However, the current protocol has

potential clinical and biological complications. Amongst them, crown discolouration,

development of resistant bacterial strains and allergic reaction to the intracanal

medication. In the case presented, a modified technique to avoid undesired crown

discolouration was applied sealing the dentinal tubules of the chamber, thus avoiding any

contact between the tri-antibiotic paste and the dentinal walls.

Key learning points

• Sealing the dentinal tubules of the chamber prevents the undesirable crown discolour-

ation produced by tri-antibiotic medication whilst maintaining the revascularization

potential of the pulp.

• Further research is warranted to seek an alternative infection control protocol capable of

preventing possible allergic reactions and development of resistant strains of bacteria, as

well as a biological material capable of inducing angiogenesis and allow a more predictable

scaffold and tissue regeneration.

doi:10.1111/j.1365-2591.2008.01467.x

Correspondence: Dr Nestor Cohenca, Department of Endodontics, University of Washington,

POB 357448, Seattle, WA 98195 7448, USA (Tel.: 1 206 543 5044; fax: 1 206 616 9085;


84 International Endodontic Journal, 42, 84–92, 2009 ª 2008 International Endodontic Journal

Keywords: crown discoloration, immature permanent tooth, open apex, pulp revascu-

larization, vital pulp therapy.

Received 1 April 2008; accepted 19 July 2008

Introduction

In 1971, Nygaard-Ostby & Hjortdal performed studies that can be considered the

forerunner of pulpal regeneration (Nygaard-Ostby & Hjortdal 1971). The studies were

aimed at determining how periodontal tissue would react, if the entire pulp was removed

from the main canal and the apical part subsequently allowed to be filled with blood.

Skoglund et al. (1978) further demonstrated that in a traumatic avulsion, blood vessels

slowly grow from the apex toward the pulp horn by replacing the necrosed pulp left behind

after the avulsion injury.

Since then, human avulsion case series (Kling et al. 1986) and controlled animal studies

(Cvek et al. 1990a,b, Ritter et al. 2004) have shown radiographic and histological evidence

of successful revascularization of immature permanent teeth after replantation. In this

situation, the necrotic uninfected pulp acts as a scaffold for the in-growth of new tissue

from the periapical area. The absence of bacteria is critical for successful revascularization

because the new tissue will stop at the level it meets bacteria in the canal space (Myers &

Fountain 1974, Yanpiset & Trope 2000). Studies to test the ability of topical antibiotics to

improve revascularization outcomes in experimental avulsions (Yanpiset & Trope 2000,

Ritter et al. 2004) have shown that topical doxycycline and minocycline can improve

radiographic and histological evidence of revascularization in immature avulsed permanent

teeth. Extrapolating from this information, it is hypothesized that once the canal infection

is controlled, it resembles the avulsed tooth that has a necrotic but sterile pulp space. The

blood clot is then introduced so as to mimic the scaffold that is in place with the ischeamic

necrotic pulp in the avulsed tooth and the access cavity is restored with a bacteria-tight

seal. However, in necrotic cases with apical periodontitis it must be recognized that the

vital tissue might not be normal pulp tissue, despite the fact that the root development

continues and dentine maturation occurs. In teeth with open apices and necrotic pulps, it

is possible that some vital pulp tissue and Hertwig’s epithelial root sheath remain. When

the canal is properly disinfected, the inflammatory process reverses and these tissues

may proliferate.

Recently, the concept of revascularization of necrotic pulps regained interest and

became an alternative conservative treatment option for young permanent teeth with

immature roots (Sato et al. 1993, Hoshino et al. 1996, Sato et al. 1996, Iwaya et al. 2001,

Banchs & Trope 2004, Windley et al. 2005, Thibodeau et al. 2007). As well stated by

Windley et al. (2005), revascularization of immature teeth with apical periodontitis

depends mainly on: (a) disinfection of the canal; (b) placement of a matrix in the canal for

tissue in-growth; and (c) a bacterial tight seal of the access opening. Since the infection of

the root canal system is considered to be polymicrobial, a combination of drugs would be

needed to treat the diverse flora. Thus, the recommended protocol combines the use of

metronidazole, ciprofloxacin and minocycline. Hoshino et al. (1996) performed a laboratory

study testing the antibacterial efficacy of these drugs alone and in combination against the

bacteria of infected dentine, infected pulps and periapical lesions. Alone, none of the

drugs resulted in complete elimination of bacteria. However, in combination, these drugs

were able to consistently sterilize all samples. In addition, a study by Sato et al. (1996)

found that this drug combination was effective in killing bacteria in the deep layers of root

canal dentine.

CASE

REPORT

85ª 2008 International Endodontic Journal International Endodontic Journal, 42, 84–92, 2009

This novel procedure exploits the full potential of the pulp for dentine deposition and

produces a stronger mature root that is better able to withstand fracture but has the

potential for clinical and biological complications. Amongst them, crown discolouration

(Windley et al. 2005), development of resistant bacterial strains (Greenstein & Polson

1998, Eickholz et al. 2002, Slots 2002) and allergic reaction to the intracanal medication

(de Paz et al. 1999, Hausermann et al. 2005, Jappe et al. 2005, Isik et al. 2007, Madsen

et al. 2007). Although coronal discolouration is not often reported in the literature in

association with the use of this tri-antibiotic medication, it is believed that the marked

discolouration may to be related to the use of minocycline. Kim et al. (2000) demonstrated

that Ledermix (Lederle Pharmaceuticals, GMBH Wolfratshausen, Germany), an intracanal

medication containing tetracycline, caused discolouration of immature teeth in a greater

degree than in mature teeth.

A case report is presented in which the pulps of bilateral mandibular premolars became

necrotic because of dens evaginatus and were revascularized using a modified novel

technique to avoid undesired crown discolouration.

Case report

An 11-year-old Asian girl was referred to the graduate endodontic clinic by her dentist for

evaluation and root canal treatment of her mandibular second premolars. The medical

history was non-contributory. A review of the dental history revealed that the patient had

sought dental care 3 months prior because of swelling and pain in the mandibular left

premolar region. The patient was prescribed penicillin VK 1000 mg daily by her general

dentist. The pain and swelling subsided within a week. Upon clinical examination, an

occlusal tubercle consistent with dens evaginatus (Fig. 1a,b) was diagnosed. Intraoral

sinus tracts, buccal to the mandibular left and right second premolars were present

(Fig. 1c,d). No caries were clinically detected. Pulp sensibility tests using 1, 1, 1,

2-tetrafluoroethane (Endo-Ice; Hygenic Corp., Akron, OH, USA) produced no response

from either mandibular second premolars whilst the adjacent mandibular first molars and

premolars responded to cold without lingering. Neither mandibular second premolars

were sensitive to percussion or palpation. Periodontal probing affirmed normal attachment

with no probing depths >3 mm and normal physiological mobility. No crown discolouration

was observed.

Radiographically both mandibular second premolars had a similar appearance, with

widened periodontal ligament space, incomplete root formation and diffuse periapical

radiolucencies 6 · 6 mm in size (Fig. 2a,b). No carious lesions were diagnosed (Fig. 2c)

and the root development appeared arrested with wide open apices in both mandibular

second premolars. A gutta-percha point was used to trace the sinus tract and a periapical

radiograph taken, demonstrating the association between the drainage and the periradic-

ular radiolucency (Fig. 2d).

Based on the results of clinical and radiographic examination, the pulpal and

periradicular diagnosis of the mandibular left and right second premolars was determined

as pulpal necrosis with chronic suppurative periradicular periodontitis. Taking into

consideration the stage of root development, the maturation of the dentinal walls and

the wide-open apices, the treatment plan included pulp revascularization of both

mandibular second premolars. After a comprehensive discussion of the risks, complica-

tions and possible outcomes of this treatment, parental consent was obtained.

Following administration of local anaesthesia, the mandibular left second premolar was

isolated with rubber dam. Under a dental-operating microscope, access preparation

was performed and a single orifice with a wide canal was revealed. No purulent exudates

or haemorrhage were observed in the chamber (Fig. 3a). Length was estimated

CASE

REPORT


radiographically using a size15 K-file. The selection of the file size was made to avoid any

damage to the canal walls. The irrigation protocol included a slow and careful irrigation of

20 mL of 6% sodium hypochlorite, 2 mm back from working length. This was followed by

a 5 mL rinse of saline and then a final irrigation of 10 mL of 2.0% chlorhexidine gluconate

(Vista Dental, Racine, WI, USA).

A modification of the current clinical protocol (Sato et al. 1996, Banchs & Trope 2004)

was established to avoid crown discolouration. This novel approach seals the dentinal

tubules of the chamber, thus avoiding any contact between the tri-antibiotic paste and

the dentinal walls. The inner surfaces of the coronal access were etched for 20 s with

35% phosphoric acid (Ultra-Etch; Ultradent, South Jordan, UT, USA) and rinsed. Bonding

agent was applied (Single Bond 3M, Minneapolis, MN, USA) to the etched surfaces and

cured for 20 s. Then, a Root Canal Projector (CJM Engineering Inc., Santa Barbara, CA,

USA) with a size 20 K-file inside the projector was placed into the prepared access to

maintain patency. The space between the projector and the coronal dentine was sealed

with flowable composite (PermaFlo DC; Ultradent, South Jordan, UT, USA) and light-

cured for 30 s (Fig. 3b). The projector was then removed by engaging it with a Hedstrom

file.

The tri-antibiotic paste was prepared immediately prior to treatment by mixing 250 mg

of Ciprofloxacin, 250 mg of Metronidazole and 250 mg of Minocycline with sterile water

(Fig. 3c). A 20G needle was set 2 mm short of working length and used to introduce the

medication into the canal using a backfill approach up to the level of the cemento-enamel

junction (CEJ) (Fig. 3d). The tooth was then temporarily sealed with a cotton pellet and

Cavit (3M ESPE, Seefeld, Germany).

One month later, the patient presented with localized swelling and pain on her

mandibular right quadrant associated with the mandibular right second premolar. The

Figure 1 (a,b) Clinical photographs showing an occlusal tubercle consistent with dens evaginatus in

both mandibular second premolars. (c,d) An intraoral localized swelling buccal to the mandibular left

and right second premolars is noted.

CASE

REPORT


swelling was visible extraorally along the lower border of the mandible in the

submandibular space, tender to palpation and non-fluctuant. The patient’s temperature

was 36.6 �C. Under local anaesthetic and rubber dam isolation, the mandibular right

second premolar was accessed for treatment. Upon access, no purulent exudate was

noted and only some minor haemorrhage. After working length was determined, the canal

was carefully irrigated with 10 mL of 6% sodium hypochlorite up to 2 mm from working

length at which time the access was sealed with a cotton pellet and Cavit as a temporary

restoration. In case of persistent pain, the patient was instructed to take ibuprofen

200 mg. Incision and drainage was not indicated, as the draining sinus tract was still

present and the swelling was non-fluctuant.

At the same appointment, it was noted that the mandibular left second premolar was

asymptomatic and was not sensitive to palpation and percussion. The sinus tract

associated with the mandibular left second premolar had healed.

Four-days later the swelling was significantly reduced, as well as the patients’ complain

of pain in the mandibular right quadrant. The sinus tract stoma associated with the

mandibular right second premolar was still present. At this appointment it was decided to

continue treatment of the mandibular left second premolar as previously planned. Under

local anaesthesia and rubber dam isolation, the tooth was re-accessed. No purulent

drainage or haemorrhage was noted upon access and the tri-antibiotic paste was removed

with 6% sodium hypochlorite with the irrigation needle tip 2 mm short of the working

(a)

(c)

(b)

(d)

Figure 2 (a,b) Radiographic examination demonstrated incomplete root formation and diffuse

periapical radiolucencies of 6 · 6 mm in size in both mandibular second premolars. (c) No carious

lesion was diagnosed. (d) A gutta-percha point was used to trace the sinus tract and a periapical

radiograph was taken, demonstrating the association between the drainage and the periradicular

radiolucency.

CASE

REPORT


length. With the root canal infection controlled, the regenerative process was initiated. A

sterile size 20 K-File was introduced 2 mm past the working length to stimulate bleeding

and create a biological scaffold for pulpal regeneration. The intracanal haemorrhage was

controlled below the CEJ by applying pressure with a sterile saline-soaked cotton pellet

until a clot was established. ProRoot grey MTA (Dentsply Tulsa Dental, Johnson City, TN,

USA) was then mixed with sterile water and carefully placed above the blood clot up to the

level of the CEJ. The access was sealed with a moist cotton pellet and Cavit. At the same

appointment, the mandibular right second premolar was anaesthetized and isolated with

rubber dam. No purulent drainage or haemorrhage was observed upon access of the

mandibular right second premolar. At this time, the same clinical protocol used to treat the

mandibular left second premolar was used on the mandibular right second premolar.

Using the same novel technique mentioned previously to avoid discolouration, flowable

composite was applied to the coronal dentine sealing the dentinal tubules preventing

contact with the tri-antibiotic paste. The tri-antibiotic dressing was placed into the canal

with a syringe set at 2 mm from the working length. The access was then sealed with a

cotton pellet and Cavit.

Two weeks following the last appointment, the patient returned asymptomatic and

without swelling or sinus tract stomas on either side of the mandible. Under local

anaesthesia and rubber dam isolation in the mandibular left quadrant, the temporary

restoration was removed from the mandibular left second premolar and the coronal

(a)

(c)

(b)

(d)

Figure 3 (a) Mandibular left second premolar was accessed under a dental-operating microscope,

without evidence of purulent exudates or haemorrhage. (b) A Root Canal Projector with a size 20 K-file

inside the projector was placed into the prepared access to maintain patency. The space between the

projector and the coronal dentine was sealed with flowable composite and cured for 30 s. (c) The tri-

antibiotic paste was prepared immediately prior to treatment and loaded in a syringe with a 20G

needle and a rubber stopper. (d) The canal was dressed using a backfill approach up to the level of the

cemento-enamel junction (CEJ) and the tooth was then temporarily sealed. Notice the flowable

composite sealing the access walls up to the level of the CEJ.

CASE

REPORT


access refined, cleaned and restored with resin-bonded composite (Prisma TPH; Dentsply

Culk, York, PA, USA).

Four weeks after the tri-antibiotic medication was placed on the mandibular right

second premolar, the patient was asymptomatic and without swelling or sinus tract

stomas on either side. Under local anaesthesia and rubber dam isolation, the tooth was

treated using the same clinical protocol used for the mandibular left second premolar.

Briefly, sodium hypochlorite irrigation was used for removal of the tri-antibiotic paste

followed by stimulation of haemorrhage, clot formation and MTA placement. The tooth

was then temporized and the patient was rescheduled for the final composite restoration

which was place 2 weeks later without incident or change in symptoms.

During the 18-month follow-up period the patient remained asymptomatic. Clinically,

both mandibular second premolars responded within normal limits to cold test using 1,

1, 1, 2-tetrafluoroethane. No tenderness to percussion or palpation was noted and the

periodontal examination revealed no pocket depths over 3 mm and normal physiological

mobility. The radiographs demonstrated evidence of periradicular bone healing and

significant root development with maturation of the dentine as compared with the

preoperative radiographs (Fig. 4a,b). Clinically, the mandibular right second premolar

teeth showed no change in shade or colour (Fig. 4c), although mandibular left second

premolar revealed a slight cervical discolouration possibly related to the use of grey

MTA (Fig. 4d).

Discussion

An immature tooth with early irreversible pulp involvement presents with thin divergent or

parallel dentinal walls. This situation creates clinical challenges in disinfection, and as a

result, affects the long-term outcome of the treatment. Traditionally, calcium hydroxide

has been used as the intra-canal medicament in apexification procedures. However,

because of its high pH it will cause necrosis of tissues that can potentially differentiate into

new pulp. Moreover, even if rendered successful, apexification procedures will leave a

short root with thin dentinal walls with a high risk of root fracture.

Revascularization of a pulp-like tissue for dentine deposition will allow further

development of the root and dentinal structure with a better long-term prognosis. As

shown in the present case, clinical and radiographic evaluation at 6-month intervals is

stressed after revascularization therapy, so as to assess pulp vitality and progression of

root development. Current vitality tests still depend on neurological stimulation and its

reliability on immature teeth is considered questionable (Fulling & Andreasen 1976, Fuss

et al. 1986). The radiographic diagnosis of periapical pathosis may also become difficult in

immature teeth because of the normal radiolucency of the developing root sheath which

occurs apically as the root matures. Comparison of root formation with the contralateral

teeth should always be performed to evaluate treatment outcome.

If crown discolouration occurs, treatment by intracoronal bleaching with sodium

perborate should be attempted. In addition, the use of white MTA instead of grey MTA

should also be considered. The modified protocol described in the present article is an

attempt to avoid the undesired crown discolouration. It also describes a safer and more

reliable technique for antibiotic dressing using a 20G needle with a backfill approach. This

novel approach prevents the undesirable crown discolouration produced by the tri-

antibiotic medication, whilst maintaining the revascularization potential of the pulp. Taking

in consideration the importance of aesthetics, this technique could be consider for all

anterior teeth in which the use of the tri-antibiotic paste is indicated for revascularization

purposes. Further research is warranted to seek an alternative infection control protocol

capable of preventing possible allergic reactions and development of resistant strains of

CASE

REPORT


bacteria, as well as a biological material capable of inducing angiogenesis and allow a more

predictable scaffold and tissue regeneration.

Disclaimer




Societies.

References

Banchs F, Trope M (2004) Revascularization of immature permanent teeth with apical periodontitis:

new treatment protocol? Journal of Endodontics 30, 196–200.

Cvek M, Cleaton-Jones P, Austin J, Lownie J, Kling M, Fatti P (1990a) Pulp revascularization in

reimplanted immature monkey incisors–predictability and the effect of antibiotic systemic

prophylaxis. Endodontics and Dental Traumatology 6, 157–69.

Cvek M, Cleaton-Jones P, Austin J, Lownie J, Kling M, Fatti P (1990b) Effect of topical application of

doxycycline on pulp revascularization and periodontal healing in reimplanted monkey incisors.

Endodontics and Dental Traumatology 6, 170–6.

Eickholz P, Kim TS, Burklin T et al. (2002) Non-surgical periodontal therapy with adjunctive topical

doxycycline: a double-blind randomized controlled multicenter study. Journal of Clinical

Periodontology 29, 108–17.

Fulling HJ, Andreasen JO (1976) Influence of maturation status and tooth type of permanent teeth

upon electrometric and thermal pulp testing. Scandinavian Journal of Dental Research 84, 286–90.

(a)

(c)

(b)

(d)

Figure 4 Eighteen-month follow-up. (a, b) Both mandibular second premolars demonstrated signif-

icant root development with maturation of the dentine. (c) Clinically, the mandibular right second

premolar presented no crown discolouration whilst (d) the mandibular left second premolar showed a

slight cervical discolouration related to the use of grey MTA.

CASE

REPORT


Fuss Z, Trowbridge H, Bender IB, Rickoff B, Sorin S (1986) Assessment of reliability of electrical and

thermal pulp testing agents. Journal of Endodontics 12, 301–5.

Greenstein G, Polson A (1998) The role of local drug delivery in the management of periodontal

diseases: a comprehensive review. Journal of Periodontology 69, 507–20.

Hausermann P, Scherer K, Weber M, Bircher AJ (2005) Ciprofloxacin-induced acute generalized

exanthematous pustulosis mimicking bullous drug eruption confirmed by a positive patch test.

Dermatology 211, 277–80.

Hoshino E, Kurihara-Ando N, Sato I et al. (1996) In-vitro antibacterial susceptibility of bacteria taken

from infected root dentine to a mixture of ciprofloxacin, metronidazole and minocycline.


Isik SR, Karakaya G, Erkin G, Kalyoncu AF (2007) Multidrug-induced erythema multiforme. Journal of

Investigational Allergology & Clinical Immunology 17, 196–8.

Iwaya SI, Ikawa M, Kubota M (2001) Revascularization of an immature permanent tooth with apical

periodontitis and sinus tract. Dental Traumatology 17, 185–7.

Jappe U, Schnuch A, Uter W (2005) Rosacea and contact allergy to cosmetics and topical

medicaments – retrospective analysis of multicentre surveillance data 1995–2002. Contact

Dermatitis 52, 96–101.

Kim ST, Abbott PV, McGinley P (2000) The effects of Ledermix paste in discolouration of immature

teeth. International Endodontic Journal 33, 233–7.

Kling M, Cvek M, Mejare I (1986) Rate and predictability of pulp revascularization in therapeutically

reimplanted permanent incisors. Endodontics and Dental Traumatology 2, 83–9.

Madsen JT, Thormann J, Kerre S, Andersen KE, Goossens A (2007) Allergic contact dermatitis to

topical metronidazole – 3 cases. Contact Dermatitis 56, 364–6.

Myers WC, Fountain SB (1974) Dental pulp regeneration aided by blood and blood substitutes after

experimentally induced periapical infection. Oral Surgery Oral Medicine Oral Pathology 37, 441–50.

Nygaard-Ostby B, Hjortdal O (1971) Tissue formation in the root canal following pulp removal.

Scandinavia Journal of Dental Research 79, 333–49.

de Paz S, Perez A, Gomez M, Trampal A, Dominguez Lazaro A (1999) Severe hypersensitivity reaction

to minocycline. Journal of Investigational Allergology Clinical Immunology 9, 403–4.

Ritter AL, Ritter AV, Murrah V, Sigurdsson A, Trope M (2004) Pulp revascularization of replanted

immature dog teeth after treatment with minocycline and doxycycline assessed by laser Doppler

flowmetry, radiography, and histology. Dental Traumatology 20, 75–84.

Sato T, Hoshino E, Uematsu H, Noda T (1993) In vitro antimicrobial susceptibility to combinations of

drugs on bacteria from carious and endodontic lesions of human deciduous teeth. Oral Microbiology

& Immunology 8, 172–6.

Sato I, Ando-Kurihara N, Kota K, Iwaku M, Hoshino E (1996) Sterilization of infected root-canal dentine

by topical application of a mixture of ciprofloxacin, metronidazole and minocycline in situ.


Skoglund A, Tronstad L, Wallenius K (1978) A microangiographic study of vascular changes in

replanted and autotransplanted teeth of young dogs. Oral Surgery Oral Medicine Oral Pathology 45,

17–28.

Slots J (2002) Selection of antimicrobial agents in periodontal therapy. Journal of Periodontal Research

37, 389–98.

Thibodeau B, Teixeira F, Yamauchi M, Caplan DJ, Trope M (2007) Pulp revascularization of immature

dog teeth with apical periodontitis. Journal of Endodontics 33, 680–9.

Windley W, Teixeira F, Levin L, Sigurdsson A, Trope M (2005) Disinfection of immature teeth with a

triple antibiotic paste. Journal of Endodontics 31, 439–43.

Yanpiset K, Trope M (2000) Pulp revascularization of replanted immature dog teeth after different

treatment methods. Endodontics and Dental Traumatology 16, 211–7.

CASE

REPORT


IEJ.01.2009

Documents

root canal biolm model

byhighconcentrated ozone

biolm cells

infected root canal

ludwigmaximilians university

munich departments

root canalcorrespondence

root canals