iej2109hjkl;

Influence of clinical experience on theradiographic determination of endodontic workinglength

O. S. Alothmani1, L. T. Friedlander2, B. D. Monteith2 & N. P. Chandler2

1Division of Endodontics, Department of Conservative Dentistry, Faculty of Dentistry, King Abdulaziz University, Jeddah, Saudi

Arabia; and 2Department of Oral Rehabilitation, School of Dentistry, University of Otago, Dunedin, New Zealand

Abstract

Alothmani OS, Friedlander LT, Monteith BD,

Chandler NP. Influence of clinical experience on the

radiographic determination of endodontic working length.

International Endodontic Journal, 46, 211216, 2013.

Aim To determine the influence of clinical experience

on the accuracy and consistency of estimation of

radiographic working length (WL) for the root canal

treatment of single-rooted teeth.

Methodology Forty conventional WL periapical

radiographs that included variations in file length

were selected. They were digitally scanned and

arranged in PowerPoint presentations on CDs. These

were distributed to three assessor groups; fourth-year

undergraduates at two stages of training (Groups 1

and 2) and endodontic postgraduates (Group 3). Par-

ticipants were asked to determine the adjustment

needed in millimetres to position the file tip at the

correct WL for each image. A gold standard file posi-

tion was provided by three experienced endodontists.

For inter-group comparison of scores, the Kruskal

Wallis, ANOVA and post hoc Bonferroni tests were

used. Evaluation of intra-examiner consistency was

with the Kappa test. To evaluate intra-group consis-

tency, the Wilcoxon signed rank test was used to

compare the frequency of weighted correct scorings.

Results File adjustments of Group 3 were signifi-

cantly more accurate than those of Group 1

(P = 0.006). The scores of Group 3 were also betterthan those of Group 2, although the difference was

not significant. When the scores of the undergraduate

groups were compared, the difference was not statisti-

cally significant. The consistency of the groups was

not affected by a 2-week pause between assessments,

and no definite pattern could be detected across any

of the groups with the Kappa test.

Conclusion Clinical experience after graduation

influenced the accuracy of estimating the adjustments

needed for correct radiographic WL of single-rooted

teeth. The most experienced group was significantly

more accurate than the other groups.

Keywords: canal preparation, endodontics, radiog-

raphy, working length.

Received 23 February 2012; accepted 4 July 2012

Introduction

Root canal treatment aims at preventing and/or

resolving apical periodontitis. The association between

infection of the root canal system and the development

of apical periodontitis has been well established

(Kakehashi et al. 1965). The entire root canal system

must be cleaned and shaped to its furthest apical limit

without violating the periapical tissue or leaving resid-

ual bacteria in the critical apical part of the canal

(Wu et al. 2000). Working length (WL) determination

has been performed by tactile sensation (Seidberg et al.

1975), use of a paper point (Marcos-Arenal et al.

2009), electronic apex locators (Gordon & Chandler

2004) and most commonly using radiographs (Ingle

Correspondence: Nicholas Chandler, School of Dentistry, Uni-

versity of Otago, PO Box 647, Dunedin 9054, New Zealand

(e-mail: [email protected]).

2012 International Endodontic Journal International Endodontic Journal, 46, 211216, 2013

doi:10.1111/j.1365-2591.2012.02109.x

211

1957). Several apical limits have been suggested for

root canal preparation, including the cementoden-

tinal junction, apical constriction, apical foramen and

radiographic apex.

The radiographic method involves estimating the

distance between the tip of a root canal file and the

radiographic apex and adjusting it to be 0.51-mm

short (Ingle 1957). Setting the WL 1-mm short of the

radiographic apex resulted in a mean radiographic

WL coronal to the apical foramen by a distance of

0.1 (2.11) mm. Furthermore, the file was at the api-cal foramen in only 16% of cases (Brunton et al.

2002). The apical constriction was precisely identified

only 1518% of the time when the WL was deter-

mined to be 1-mm short of the radiographic apex

(Pratten & McDonald 1996, Vieyra et al. 2010). The

file tip was coronal to the cemento-dentinal junction,

apical constriction and apical foramen when the WL

was adjusted to 0.5-mm short of the radiographic

apex (Hassanien et al. 2008). Despite an acceptable

radiographic WL appearance, 1925% of measure-

ments have been shown to be beyond the apical fora-

men (Chunn et al. 1981, ElAyouti et al. 2001, Welk

et al. 2003).

The accuracy of the radiographic method is influ-

enced by factors including the type of teeth (ElAyouti

et al. 2001, Chen et al. 2011), position of the apical

foramen (Olson et al. 1991, Blaskovic-Subat et al.

1992), curvature of the root canal (Kim-Park et al.

2003) and film speed and radiographic modality, that

is, conventional versus digital (Ellingsen et al. 1995a,b).

Observer variability (Cox et al. 1991) and viewing

conditions (Orafi et al. 2010) have also been sug-

gested as potential factors. What is not clear is the

impact clinical experience may have on the ability to

assess WL accurately.

Cox et al. (1991) compared the agreement of nine

observers: three endodontists, three oral radiologists

and three general practitioners concerning the adjust-

ments needed for optimum radiographic WL. The

assessors were asked to identify the distance to place

a file tip 0.5-mm short of radiographic apex, or short

of where the apical foramen was perceived. Although

the assessors were given identical instructions, their

implementation varied, with excellent agreement,

defined as 0.5-mm variation from the actual adjust-ment, achieved for only 68% of the radiographs (Cox

et al. 1991). The variation could be a result of their

clinical experience. A recent study investigated this

and found that the agreement of endodontists, restor-

ative dentists, general practitioners, endodontic

postgraduates and undergraduate students was high,

indicating that difference in experience did not influ-

ence their interpretation (Orafi et al. 2010). The

radiographs used were of extracted teeth and were

taken under ideal exposure conditions in vitro. A com-

bination of electronic and radiographic methods is

recommended for WL determination by the European

Society of Endodontology (2006), so further explora-

tion of the influence of clinical experience on the

interpretation of radiographic WL using clinical

images would be valuable.

The objective of this study was to determine the

influence of clinical experience on the accuracy and

consistency of estimation of radiographic WL adjust-

ments indicated for the root canal treatment of single-

rooted teeth. The null hypothesis was that clinical

experience would not influence adjustments of length.

Materials and methods

Ethical approval was granted to access patient records,

and 40 WL periapical radiographs taken during root

canal treatment of single-rooted teeth were selected.

Variations in file length in relation to the radiographic

apex were included. The images were exposed by dif-

ferent operators using the paralleling technique and a

Rinn XCP film holder with beam-guiding ring (Dents-

ply Rinn, Elgin, IL, USA). E- and F-speed films had

been used (Eastman Kodak Co, Rochester, NY, USA).

The X-ray source was a MINRAY unit (Soredex, Tuu-

sula, Finland) set at 70 kV, 7 mA with an exposure

range of 0.160.25 s depending on the tooth, and

films were automatically processed.

Radiographs were fitted into transparency mounts,

scanned (Nikon LS-5000 scanner; Nikon Corp, Tokyo,

Japan) and saved in JPEG format. All images had

dimensions of 1640 9 1065 pixels with 4000 dpi

resolution. The 40 images were arranged in a Power-

Point presentation (Microsoft Corp, Redmond, WA,

USA) and distributed to assessor groups on CDs.

Assessment I

The assessor groups comprised:

Group 1: Thirty-five 4th-year undergraduate den-

tistry students in the first 6 months of the aca-

demic year (the least experienced group). The

students had received all didactic teaching and

completed a pre-clinical course, which involved

root filling extracted teeth. They had very limited

clinical experience.

Radiographic working length Alothmani et al.

2012 International Endodontic JournalInternational Endodontic Journal, 46, 211216, 2013212

Group 2: Thirty-five 4th-year undergraduate den-

tistry students who assessed the images at years

end. They had 6-month endodontic patient care

experience.

Group 3: Ten endodontic postgraduates (the most

experienced group). They had 310-year clinical

experience since graduation.

Participants were coded to preserve anonymity and

minimize bias. They were given instructions and

viewed the CD on their own computers. They were

asked to determine the adjustment needed (to within

0.5 mm) to position the file tip at the correct WL for

each image. All the assessors had been taught during

their undergraduate training that the ideal file tip

position was on average 1 mm from the radiographic

apex of the tooth.

A gold standard was provided by three experienced

endodontists who had been on the New Zealand spe-

cialists register for 6, 9 and 19 years. Their evalua-

tions were carried out separately, and any

disagreements discussed to reach a consensus.

Assessment II

Two weeks later, intra-observer agreement was deter-

mined with a second presentation containing half of

the original images. This was only carried out for

those individuals who returned the first assessment.

Statistical analysis

For inter-group comparison of scores, the Kruskal

Wallis, ANOVA and post hoc Bonferroni tests were

performed with significance set at 0.05. Evaluation of

individuals intra-examiner consistency was deter-

mined by the Kappa test. To evaluate intra-group

consistency, the Wilcoxon signed rank test was used

to compare the frequencies of weighted correct scor-

ings (a discrepancy of 0.5 mm from the gold stan-dard was considered accurate) for assessment I and II

to reveal any significant changes as a result of the 2-

week lag.

Results

Completed evaluation sheets of both assessments were

received from 13/35 (37.1%) members of Group 1,

18/35 (51.4%) members of Group 2 and from all

members of Group 3.

Inter-group consistency

Comparing the scores of the groups using the Kruskal

Wallis, ANOVA and post hoc Bonferroni tests showed

that the scores of Group 3 were significantly more

accurate than those of Group 1 (P = 0.006). Thescores of Group 3 were also more accurate than

Group 2, although the difference was not statistically

significant (P = 0.093). Similarly, when the scores ofthe two undergraduate groups were compared, the

difference was not significant, although Group 2 was

more accurate (P = 0.509).

Intra-examiner consistency

Regarding Kappa values, no definite pattern could be

detected across any of the groups (Table 1).

Intra-group consistency

There was no significant difference between the fre-

quency of weighted correct scores for assessments I

and II for each of the three groups when the Wilco-

xon signed rank test was applied.

Discussion

In the study, evaluators with different clinical experi-

ence assessed 40 indirectly digitized radiographs of

single-rooted teeth to indicate the amount of adjust-

ment needed to achieve correct WL. The only signifi-

cant difference was found when the adjustments of

the most experienced group (Group 3) were compared

to those of the least experienced group (Group 1).

Regarding intra-group consistency, the three groups

Table 1 Kappa scores for Assessments I and II

Kappa scores (range) Frequency of

higher Kappa

score for

Assessment I

Frequency of

higher Kappa

score for

Assessment II

Frequency of

equal Kappa

scores for both

assessmentsAssessment I Assessment II

Group 1 0.070.30 0.050.47 3 8 2Group 2 0.020.47 0.060.49 8 9 1Group 3 0.130.46 0.160.41 3 5 2

Alothmani et al. Radiographic working length

2012 International Endodontic Journal International Endodontic Journal, 46, 211216, 2013 213

were consistent with themselves with no significant

difference between the frequency of the weighted cor-

rect scores between the first and second assessments.

Individuals intra-examiner consistency varied, with

no definite pattern being observed.

To match clinical settings as closely as possible, it

was decided to use periapical WL radiographs of good

quality exposed during routine clinical treatment.

Films had been exposed using the paralleling tech-

nique as recommended (Forsberg 1987a,b,c). Images

of single-rooted teeth were chosen for simplification of

interpretation; length assessment for multi-rooted

teeth is more complex because of superimposition by,

for example, the zygomatic arch (Tamse et al. 1980).

Radiographic WL determination might also be more

accurate for anterior teeth because of the reduced

possibility of apical foramen deviation compared with

posterior teeth (Williams et al. 2006). The frequency

of WL assessments leading to instrumentation beyond

the apical foramen in pre-molars and molars has been

shown to be 51% and 22%, respectively, whilst this

did not take place in anterior teeth (ElAyouti et al.

2001).

Acquisition of high-quality scanned images was of

prime importance, so the digital resolution was set at

4000 dpi with dimensions of 1640 9 1065 pixels.

The minimum image dimensions considered necessary

for WL determination are 800 9 600 pixels (Luos-

tarinen 2004). One in vitro study reported that the tip

of a size 6 hand file was better visualized on conven-

tional E-speed films than their scanned counterparts

(Fuge et al. 1998), but scanning was with a flat-bed

scanner rather than the high-quality film/slide instru-

ment used in the present study. Another study using

a film/slide scanner reported that the clarity of the tip

of a size 8 file on scanned images was not superior to

the conventional images (Goga et al. 2004). In the

present study, the clarity of file tips was ensured in all

the images, and none of the assessors reported that

the instruments could not be seen.

Volunteer participants with a variable range of clin-

ical experience were involved and great efforts made

to ensure their participation. The principle investiga-

tor followed up the students by emails, notes and

face-to-face reminders. Whilst all the endodontic post-

graduate students took part, acquiring completed

assessment sheets from the undergraduate students

was far more problematic leading to unequal group

sizes.

Fortuitously, all 10 postgraduates were graduates

of the Schools BDS programme and had received

comparable WL instruction to their juniors. Evalua-

tion of images was carried out on participants own

computers with different monitor screen size and reso-

lution. Resolution might influence the interpretation

of digital images (Goga et al. 2004). The images were

probably viewed at different times of the day, with

variable mental status and under different ambient

lighting. The influence of the mental status of asses-

sors on radiographic interpretation has been high-

lighted (Goldman et al. 1972, Tidmarsh 1987).

According to Ellingsen et al. (1995a), digital images

should be viewed in a dark room, but another study

reported that this enhanced the speed of identifying

the file tip position but not its accuracy (Heo et al.

2008).

No specific definition for correct WL was given to

the assessors. This ensured resemblance to the clinical

situation, where practitioners may differ in their cho-

sen end-point in relation to the radiographic apex.

Further, the purpose of the study was to correlate the

clinical experience of the evaluators with their inter-

pretation of radiographic WL. Such an evaluation

might not be feasible when a defined position for WL

is provided (Cox et al. 1991).

The lack of significant difference in intra-group con-

sistency validated adopting the 2-week period

between assessments. When the individuals Kappa

values for assessment I and II were compared, no defi-

nite pattern emerged. The values of some evaluators

were higher in assessment I whilst the values in

assessment II were higher for others. The Kappa val-

ues for a few evaluators were unchanged (Table 1).

Reporting Kappa values was chosen to indicate the

agreement of a subject and taking into account the

possibility of a chance concord; an advantage lacking

when per cent agreement is reported (Cohen 1960).

The significantly higher accuracy of Group 3 com-

pared with Group 1 indicates that radiographic WL

determination requires training and practice to

achieve more accurate results. Although not signifi-

cant, Group 3 scores were also more accurate than

those of Group 2. This probably suggests that the

brief experience gained during the 6-month academic

semester was sufficient to fill the gap that existed

between the most experienced group and the least

experienced group.

Cox et al. (1991) were the first to point out the

variability in radiographic WL determination amongst

evaluators. However, they did not relate their findings

to the experience of the assessors although their

study included three groups with different clinical



backgrounds. A recent study reported that the diver-

sity in clinical experience of evaluators did not result

in a significant difference in their estimation of root

canal file or tooth length (Orafi et al. 2010). This con-

tradicts the present findings, which show that the

most experienced group was significantly more accu-

rate than the least experienced group. Orafi et al.

(2010) used conventional radiographs exposed in vi-

tro using individual teeth mounted in acrylic blocks,

and the assessors were asked to determine file length

using a ruler. The current study used indirectly digi-

tized radiographs exposed in vivo, and without the use

of a measuring tool. The assessors were asked to

determine the adjustment needed (not file length) to

obtain the correct WL.

Conclusions

The clinical experience of the assessors influenced

their accuracy of estimating the adjustment needed

for radiographic WL of single-rooted teeth. The esti-

mations of the most experienced group were signifi-

cantly more accurate than those of the least

experienced group. Meanwhile, Kappa values differed

between assessments but no definite pattern could be

detected. The performance of all three groups was not

affected by a 2-week pause between assessments.

References

Blaskovic-Subat V, Maricic B, Sutalo J (1992) Asymmetry of

the root canal foramen. International Endodontic Journal

25, 15864.

Brunton PA, Abdeen D, Macfarlane TV (2002) The effect of

an apex locator on exposure to radiation during endodon-

tic therapy. Journal of Endodontics 28, 5246.

Chen E, Kaing S, Mohan H, Ting S-Y, Wu J, Parashos P

(2011) An ex vivo comparison of electronic apex locator

teaching models. Journal of Endodontics 37, 114751.

Chunn CB, Zardiackas LD, Menke RA (1981) In vivo root

canal length determination using the Forameter. Journal of

Endodontics 7, 51520.

Cohen J (1960) A coefficient of agreement for nominal scales.

Educational and Psychological Measurement 20, 3746.

Cox VS, Brown JCE, Bricker SL, Newton CW (1991) Radio-

graphic interpretation of endodontic file length. Oral Sur-

gery, Oral Medicine, Oral Pathology 72, 3404.

ElAyouti A, Weiger R, Lost C (2001) Frequency of overin-

strumentation with an acceptable radiographic working

length. Journal of Endodontics 27, 4952.

Ellingsen MA, Harrington GW, Hollender LG (1995a) Radio-

visiography versus conventional radiography for detection of

small instruments in endodontic length determination. Part

1. In vitro evaluation. Journal of Endodontics 21, 32631.

Ellingsen MA, Hollender LG, Harrington GW (1995b) Radio-

visiography versus conventional radiography for detection

of small instruments in endodontic length determination

II. In vivo evaluation. Journal of Endodontics 21, 51620.

European Society of Endodontology (2006) Quality guide-

lines for endodontic treatment: consensus report of the

European Society of Endodontology. International Endodon-

tic Journal 39, 92130.

Forsberg J (1987a) A comparison of the paralleling and

bisecting-angle radiographic techniques in endodontics.

International Endodontic Journal 20, 17782.

Forsberg J (1987b) Estimation of the root filling length with

the paralleling and bisecting-angle techniques performed

by undergraduate students. International Endodontic Journal

20, 2826.

Forsberg J (1987c) Radiographic reproduction of endodontic

working length comparing the paralleling and the

bisecting-angle techniques. Oral Surgery, Oral Medicine,

Oral Pathology 64, 35360.

Fuge KN, Stuck AMC, Love RM (1998) A comparison of digi-

tally scanned radiographs with conventional film for the

detection of small endodontic instruments. International

Endodontic Journal 31, 1236.

Goga R, Chandler NP, Love RM (2004) Clarity and diagnos-

tic quality of digitized conventional intraoral radiographs.

Dentomaxillofacial Radiology 33, 1037.

Goldman M, Pearson AH, Darzenta N (1972) Endodontic

successWhos reading the radiograph? Oral Surgery, Oral

Medicine, Oral Pathology 33, 4327.

Gordon MPJ, Chandler NP (2004) Electronic apex locators.

International Endodontic Journal 37, 42537.

Hassanien EE, Hashem A, Chalfin H (2008) Histomorpho-

metric study of the root apex of mandibular premolar

teeth: an attempt to correlate working length measured

with electronic and radiograph methods to various ana-

tomic positions in the apical portion of the canal. Journal

of Endodontics 34, 40812.

Heo M-S, Han D-H, An B-M et al. (2008) Effect of ambient

light and bit depth of digital radiograph on observer per-

formance in determination of endodontic file positioning.

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

Endodontology 105, 23944.

Ingle JI (1957) Endodontic instruments and instrumentation.

Dental Clinics of North America 1, 80522.

Kakehashi S, Stanley HR, Fitzgerald RJ (1965) The effects of

surgical exposures of dental pulps in germ-free and con-

ventional laboratory rats. Oral Surgery, Oral Medicine, Oral

Pathology 20, 3409.

Kim-Park MA, Baughan LW, Hartwell GR (2003) Working

length determination in palatal roots of maxillary molars.

Journal of Endodontics 29, 5861.

Luostarinen T (2004) Digitized conventional radiograph.

Dentomaxillofacial Radiology 33, 4201.

Alothmani et al. Radiographic working length

2012 International Endodontic Journal International Endodontic Journal, 46, 211216, 2013 215

Marcos-Arenal JL, Rivera EM, Caplan DJ, Trope M (2009)

Evaluating the paper point technique for locating the api-

cal foramen after canal preparation. Oral Surgery, Oral

Medicine, Oral Pathology, Oral Radiology, Endodontology

108, e1015.

Olson AK, Goering AC, Cavataio RE, Luciano J (1991) The

ability of the radiograph to determine the location of the

apical foramen. International Endodontic Journal 24, 2835.

Orafi I, Worthington HV, Qualtrough AJE, Rushton VE

(2010) The impact of different viewing conditions on

radiological file and working length measurement. Interna-

tional Endodontic Journal 43, 6007.

Pratten DH, McDonald NJ (1996) Comparison of radio-

graphic and electronic working lengths. Journal of Endodon-

tics 22, 1736.

Seidberg BH, Alibrandi BV, Fine H, Logue B (1975) Clinical

investigation of measuring working lengths of root canals

with an electronic device and with digital-tactile sense.

Journal of the American Dental Association 90, 37987.

Tamse A, Kaffe I, Fishel D (1980) Zygomatic arch interfer-

ence with correct radiographic diagnosis in maxillary

molar endodontics. Oral Surgery, Oral Medicine, Oral

Pathology 50, 5635.

Tidmarsh BG (1987) Radiographic interpretation of end-

odontic lesions- a shadow of reality. International Dental

Journal 37, 105.

Vieyra JP, Acosta J, Mondaca JM (2010) Comparison of

working length determination with radiographs and two

electronic apex locators. International Endodontic Journal

43, 1620.

Welk AR, Baumgartner JC, Marshall JG (2003) An in vivo

comparison of two frequency-based electronic apex loca-

tors. Journal of Endodontics 29, 497500.

Williams CB, Joyce AP, Roberts S (2006) A comparison

between in vivo radiographic working length determina-

tion and measurement after extraction. Journal of Endodon-

tics 32, 6247.

Wu M-K, Wesselink PR, Walton RE (2000) Apical terminus

location of root canal treatment procedures. Oral Surgery,

Oral Medicine, Oral Pathology, Oral Radiology, Endodontology

89, 99103.



iej2109hjkl;

Documents

Transcript of iej2109hjkl;