cone beam pai estrela et al 2008

A New Periapical Index Based on Cone BeamComputed TomographyCarlos Estrela, DDS, MSc, PhD,* Mike Reis Bueno, DDS, MSc,†

Bruno Correa Azevedo, DDS, MSc,‡ José Ribamar Azevedo, DDS,§ andJesus Djalma Pécora, DDS, MSc, PhD!

AbstractThe purpose of this study was to evaluate a newperiapical index based on cone beam computed tomog-raphy (CBCT) for identification of apical periodontitis(AP). The periapical index proposed in this study (CBCT-PAI) was developed on the basis of criteria establishedfrom measurements corresponding to periapical radi-olucency interpreted on CBCT scans. Radiolucent im-ages suggestive of periapical lesions were measured byusing the working tools of Planimp software on CBCTscans in 3 dimensions: buccopalatal, mesiodistal, anddiagonal. The CBCTPAI was determined by the largestlesion extension. A 6-point (0–5) scoring system wasused with 2 additional variables, expansion of corticalbone and destruction of cortical bone. A total of 1014images (periapical radiographs and CBCT scans) origi-nally taken from 596 patients were evaluated by 3observers by using the CBCTPAI criteria. AP was iden-tified in 39.5% and 60.9% of cases by radiography andCBCT, respectively (P ! .01). The CBCTPAI offers anaccurate diagnostic method for use with high-resolu-tion images, which can reduce the incidence of false-negative diagnosis, minimize observer interference, andincrease the reliability of epidemiologic studies, espe-cially those referring to AP prevalence and severity.(J Endod 2008;34:1325–1331)

Key WordsApical periodontitis, cone beam computed tomography,diagnostic imaging, endodontic diagnosis, radiography

Periapical radiography is an essential resource in endodontic diagnosis, because itoffers important evidence on the progression, regression, and persistence of apical

periodontitis (AP) (1, 2). It is known that periapical radiolucencies might not be visibleradiographically, although they exist clinically (3–5). Moreover, radiographs are 2-di-mensional representations of 3-dimensional structures, and certain clinical and bio-logic features might not be reflected in radiographic changes. Clinical and radiologiccriteria are frequently used to assess the status of endodontic treatment and itscorrelation with AP (6 –14), but morphologic variations, surrounding bone den-sity, x-ray angulations, and radiographic contrast can influence radiographic in-terpretation (12, 13).

With the great technological advances of recent years, new imaging modalitieshave been added to dental radiology as viable diagnostic tools, namely digital radiog-raphy, densitometry methods, cone beam computed tomography (CBCT), magneticresonance imaging, ultrasound, nuclear techniques (13, 15–23), providing detailedhigh-resolution images of oral structures and permitting early detection of bone lesions.

CBCT has been used for several clinical and investigational purposes in endodon-tics (15, 16, 18 –20). According to a recent study (16), the specific endodontic appli-cations of cone beam volumetric tomography (CBVT) are being identified as this tech-nology becomes more prevalent, but its potential indications include diagnosis ofpathosis from endodontic and nonendodontic origins, assessment of root canal mor-phology, evaluation of root and alveolar fractures, analysis of external and internal rootresorption and invasive cervical resorption, and presurgical planning in root-end sur-geries. Scarfe et al. (21) have stated that important innovations of imaging systemsinvolve the change from analog to digital imaging and advances in imaging theory andvolume-acquisition data, allowing detailed 3-dimensional imaging. Investigations (22,23) have demonstrated that a cyst could be distinguished from periapical granulomasby CBCT because it shows a marked difference in density between the content of the cystcavity and granulomatous tissue, thus favoring a noninvasive diagnosis. At this point intime, scientific consensus has been reached to the fact that AP is correctly detected byhistologic analysis (24).

Previous studies (1, 8, 10) have referred to the periapical index (PAI) as a scoringsystem for radiographic assessment of AP. The PAI represents an ordinal scale of 5scores ranging from no disease to severe periodontitis with exacerbating features and isbased on reference radiographs with confirmed histologic diagnosis originally pub-lished by Brynolf (8). Ørstavik et al. (1) applied the PAI to both clinical trials andepidemiologic surveys and might be transformed into success and failure criteria bydefining cutoff points on the scale for a dichotomous outcome assessment (13).

Given the limitations of conventional radiography for detection of AP and theavailability of new emerging 3-dimensional imaging modalities, the development of newperiapical indices seems to be a necessity. The purpose of this study was to evaluate anew PAI based on CBCT for identification of AP.

Material and MethodsCone Beam Computed Tomography Periapical Index

The Cone Beam Computed Tomography Periapical Index (CBCTPAI) proposed inthis study was developed on the basis of criteria established from measurements cor-responding to periapical radiolucency interpreted on CBCT scans. The sizes of radiolu-

From the *Federal University of Goiás, Goiânia, GO, Brazil;†University of Cuiabá, Cuiabá, MT, Brazil; ‡Department ofDental Diagnostic Science, University of Texas, San Antonio,Texas; §Dental and Radiological Institute of Brasília, Brasília,DF, Brazil; and !University of São Paulo, Ribeirão Preto, SP,Brazil.

Address requests for reprints to Prof Dr Carlos Estrela, Centrode Ensino e Pesquisa Odontológica do Brazil (CEPOBRAS), RuaC-245, Quadra 546, Lote 9, Jardim América, CEP: 74.290-200,Goiânia, GO, Brazil. E-mail address: [email protected]/$0 - see front matter

Copyright © 2008 American Association of Endodontists.doi:10.1016/j.joen.2008.08.013

Clinical Research

JOE — Volume 34, Number 11, November 2008 Computed Tomography Periapical Index 1325

mailto:[email protected]

cent images suggestive of periapical lesions were delimited and mea-sured by using the working tools of Planimp software (CDT Computing,Cuiabá, MT, Brazil) on CBCT scans in 3 dimensions: buccopalatal, me-siodistal, and diagonal (Fig. 1). The CBCTPAI was determined by the

Figure 1. Clinical case of mandibular molar showing the axial (A), sagittal (B), and coronal (C) planes. The CBCTPAI was determined by the largest extension of thelesion.

Figure 2. Schematic representation of molars CBCTPAI. Figure 3. Schematic representation of premolars CBCTPAI.

TABLE 1. Cone Beam Computed Tomography Periapical Index Scores

Score Quantitative Bone Alterations in MineralStructures

0 Intact periapical bone structures1 Diameter of periapical radiolucency " 0.5–1 mm2 Diameter of periapical radiolucency " 1–2 mm3 Diameter of periapical radiolucency " 2–4 mm4 Diameter of periapical radiolucency " 4–8 mm5 Diameter of periapical radiolucency " 8 mm

Score (n)# E*

Expansion of periapical cortical bone

Score (n)# D*

Destruction of periapical cortical bone

*The variables E (expansion of cortical bone) and D (destruction of cortical bone) were added to eachscore, if either of these conditions was detected in the CBCT analysis.

Clinical Research

1326 Estrela et al. JOE — Volume 34, Number 11, November 2008

largest extension of the lesion. A 6-point (0 –5) scoring system wasused. In addition, considering that CBCT provides 3-dimensional im-ages, with depth being added as a new plane of analysis in relation to2-dimensional radiography, 2 variables were included in the scoringsystem as appropriate, expansion of cortical bone (E) and destructionof cortical bone (D) (Table 1 and Figs. 2–7).

PatientsOne thousand fourteen images (periapical radiographs and CBCT

scans) originally taken from 596 patients (241 men and 355 women;mean age, 54 $ 17 years) between May 2004 and August 2006 wereselected from the Dental and Radiological Institute of Brasília (IORB,Brasília, DF, Brazil) database. All patients had 1 or more teeth withhistory of endodontic treatment. Some of the teeth in question radio-graphically displayed an AP. The involved teeth were mandibular molars(n % 126), maxillary molars (n % 203), mandibular premolars (n %183), maxillary premolars (n % 226), mandibular canines (n % 11),maxillary canines (n % 99), mandibular incisors (n % 13), and max-illary incisors (n % 153).

The study design was independently reviewed and approved by theInstitutional Ethics in Research Committee.

Imaging Methods and EvaluationThe periapical radiographs were taken with Max S-1 x-ray equip-

ment (J. Morita Mfg Corp, Kyoto, Japan) with 0.8 mm & 0.8 mm tubefocal spot and with Kodak Insight film-E (Eastman Kodak Co, Rochester,NY) according to the parallel radiographic technique. All films wereprocessed in an automatic processor and developed by using standard-ized methods. CBCT images were obtained with 3D Accuitomo XYZ SliceView Tomograph (model MCT-1; J. Morita Mfg Corp) voxel size of0.125 & 0.125 & 0.125 mm, 12 or 8 bits. Images were examined byusing specific software (3D Tomo X version 1.0.51) in a PC workstationrunning under Microsoft Windows XP professional SP-1 (MicrosoftCorp, Redmond, WA).

Three calibrated blinded observers evaluated all digital images byusing the CBCTPAI criteria described in Table 1. The level of interob-server agreement was assessed by kappa statistics in 10% of the sample(25). Data were analyzed statistically by the !2 test to determine signif-icant differences among the imaging methods for detection of AP. Sig-nificance level was set at " % 1%.

ResultsTable 2 shows the prevalence of AP identified by periapical radi-

ography and CBCT by using the CBCTPAI. AP was detected in 39.5% and60.9% of cases by periapical radiography and CBCT, respectively (P !

Figure 5. Schematic representation of incisors CBCTPAI.

Figure 4. Schematic representation of canines CBCTPAI.

Clinical Research


.01). Table 3 presents the results of the diagnostic imaging tests by usingCBCTPAI kappa values for interobserver agreement, considering theCBCTPAI scores ranged from 0.86 – 0.96 for periapical radiographsand CBCT scans. Figs. 2–7 show the schematic representation and clin-ical cases of groups of mandibular and maxillary teeth corresponding toCBCTPAI.

DiscussionCBCT was more accurate than periapical radiography for AP

diagnosis. AP was identified in nearly 40% of the cases by usingperiapical radiographs and in almost 61% of the cases by usingCBCT scans (Table 2).

The accuracy of CBCT scans compared with periapical radio-graphic images is in accordance with the findings of previous studies (5,16, 18, 19, 21–23, 26). Lofthag-Hansen et al. (26) compared intraoralperiapical radiography and a 3-dimensional imaging system (3D Accui-tomo) for the diagnosis of apical pathology in 36 patients (46 teeth).

When both diagnostic methods were analyzed by all observers, theyagreed that the CBCT images provided clinically relevant additional in-formation not found in the periapical films. The capacity of computedtomography to evaluate a region of interest in 3 dimensions might ben-efit both novice and experienced clinicians alike. The advantages in-clude increased accuracy, higher resolution, scan-time reduction, andlower radiation dose (16).

The use of conventional radiography for detection of AP should bedone with care because of the great possibility of false-negative diagno-sis. The benefits of using CBCT in endodontics refer to its high accuracyin detecting periapical lesions even in its earliest stages and aiding indifferential diagnosis as a noninvasive technique (5).

Simon et al. (23) compared the differential diagnosis of largeperiapical lesions (granuloma versus cyst) with traditional biopsy byusing CBCT. Seventeen large periapical radiolucencies (equal to orgreater than 1 & 1 cm) were scanned to determine densities, and apreoperative CBCT diagnosis was made. The lesions were then removed

Figure 6. Clinical cases of maxillary incisors CBCTPAI showing all scores used and 2 additional variables (expansion of cortical bone and destruction of corticalbone).

Clinical Research


surgically and sent for histopathologic examination. In 13 of 17 cases,the biopsy report and CBCT diagnosis coincided. In 4 of 17 cases, theCBCT read cyst, whereas the oral pathologist’s diagnosis was chronicAP. These results suggest that CBCT might provide a faster method todifferentially diagnose a solid from a fluid-filled lesion or cavity, withoutinvasive surgery and/or waiting a long time to see whether nonsurgicaltherapy is effective.

The CBCTPAI proposed hereby has some advantages for clinicalapplications. CBCTPAI scores are calculated by analysis of the lesion in3 dimensions, with CT slices being obtained in mesiodistal, buccopala-tal, and diagonal directions. The measurement of lesion depth contrib-utes significantly to the diagnosis and consequently to improve caseprognosis. The addition of the variables expansion of cortical bone anddestruction of cortical bone to CBCTPAI scoring system permits the analysisof 2 possible sequels to AP that might be missed by periapical radiography.Detection of these conditions will alter the diagnostic hypothesis and thetreatment plan. The goal of this new index is therefore to offer a methodbased on the interpretation of high-resolution images that can provide amore precise measurement of AP extension, minimizing observer interfer-ence and increasing the reliability of research results.

The PAI as indicated by Ørstavik et al. (1) ranges from health tosevere periodontitis on the basis of the interpretation of 2-dimensionalradiographic images, whereas CBCTPAI scores have been establishedaccording to the interpretation of 3-dimensional CBCT scans. These 2 peri-

TABLE 2. AP Prevalence in Endodontically Treated Teeth as Determined byPeriapical Radiography and CBCT, by Using CBCTPAI (n % 1014)

PeriapicalRadiography CBCT P

Value*Presence of AP 401 (39.5%) 618 (60.9%) !.001Absence of AP 613 (60.5%) 396 (39.1%)

AP, apical periodontitis; CBCT, cone beam computed tomography.*!2 test.

Figure 7. Clinical cases of maxillary molars CBCTPAI showing all scores used and 2 additional variables (expansion of cortical bone and destruction of corticalbone).

Clinical Research


apical indices thus have different scoring systems as a result of the charac-teristics of each target image (conventional periapical radiograph or CBCTscan). The scope of the present study is to evaluate a new PAI that is basedon an imaging modality that allows detection of lesions that are not visibleradiographically. Considering the clinical history, the relation betweenhealth and disease in CBCTPAI can be begun by stage 1, without detection ofcortical bone expansion or destruction of the cortical bone.

The limitations of periapical radiography to identify AP support theneed to review the epidemiologic studies conducted in different popu-lations worldwide. A considerable discrepancy among the imagingmethods used to diagnose AP, especially with a new baseline value,certainly might reduce the influence of radiographic interpretation andthe possibility of false-negative diagnosis.

In the present study, the kappa values for CBCTPAI scores rangedfrom 0.86 – 0.96 for periapical radiographs and CBCT scans, whichindicate that a very good interobserver agreement was reached. Highkappa values (0.80 – 0.95 range) have also been observed in previousstudies that examined periapical radiographs (27, 28).

In spite of the dental imaging technique, care should be taken toavoid misinterpretation. The presence of intracanal metallic posts, forexample, might lead to equivocated interpretations as a result of artifactformation in CBCT images. Metallic objects can be present in either thetooth of interest or an adjacent one and hinder the analysis of CBCTimages (26), although in current days the influence of this artifact hasbeen reduced. This artifact has been found to be closely related to thetype of tissue or object (ie, computed tomography value) and the x-rayenergy applied (29).

Under the tested conditions and within the limitations of this study,it might be concluded that AP detection was considerably higher withCBCT than with periapical radiography. The PAI proposed in this study(CBCTPAI) offers an accurate diagnostic method for use with high-resolution images, which can reduce the incidence of false-negativediagnosis, minimize observer interference, and increase the reliabilityof epidemiologic studies, especially those referring to AP prevalenceand severity.

References1. Ørstavik D, Kerekes K, Eriksen HM. The periapical index: a scoring system for

radiographic assessment of apical periodontitis. Endod Dent Traumatol 1986;2:20 – 4.

2. Nair PNR, Sjögren U, Figdor D, Sundqvist G. Persistent periapical radiolucencies ofroot-filled human teeth, failed endodontic treatments, and periapical scars. OralSurg Oral Med Oral Pathol Oral Radiol Endod 1999;87:617–27.

3. Bender IB, Seltzer S. Roentgenographic and direct observation of experimental le-sions in bone I. J Am Dent Assoc 1961;62:152– 60.

4. Bender IB, Seltzer S. Roentgenographic and direct observation of experimental le-sions in bone II. J Am Dent Assoc 1961;62:708 –16.

5. Estrela C, Bueno MR, Leles CR, Azevedo BC, Azevedo JR. Accuracy of cone beamcomputed tomography, panoramic and periapical radiographic for the detection ofapical periodontitis. J Endod 2008;34:273–9.

6. Strindberg LZ. The dependence of the results of pulp therapy on certain factors: ananalytical study based on radiographic and clinical follow-up examinations. ActaOdontol Scand 1956;14:1–175.

7. Grossman LI, Sheprd LI, Pearson LA. Roentgenologic and clinical evaluation of end-odontically treated teeth. Oral Surg Oral Med Oral Pathol 1964;17:368 –74.

8. Brynolf I. A histologic and roentgenologic study of the periapical region of humanupper incisors. Odontol Revy 1967;18:1–176.

9. Goldman M, Pearson AH, Darzenta N. Reliability of radiographic interpretations. OralSurg Oral Med Oral Pathol 1974;38:287–93.

10. Reit C, Grøndahl HG. Application of statistical decision theory to radiographic diag-nosis of endodontically treated teeth. Scand J Dent Res 1983;9:213– 8.

11. Halse A, Molven O. A strategy for the diagnosis of periapical pathosis. J Endod1986;12:534 – 8.

12. Molven O, Halse A, Fristad I. Long-term reliability and observer comparisons in theradiographic diagnosis of periapical disease. Int Endod J 2002;35:142–7.

13. Huumonen S, Ørstavik D. Radiological aspects of apical periodontitis. Endod Topics2002;1:3–25.

14. Estrela C, Leles CR, Hollanda ACB, Moura MS, Pécora JD. Prevalence and risk factorsof apical periodontitis in endodontically treated teeth in a selected population ofBrazilian adults. Braz Dent J 2008;19:34 –9.

15. Arai Y, Tammisalo E, Iwai K, Hashimoto K, Shinoda K. Development of a compactcomputed tomographic apparatus for dental use. Dent Maxillofac Radiol 1999;28:245– 8.

16. Cotton TP, Geisler TM, Holden DT, Schwartz SA, Schindler WG. Endodontic applica-tions of cone beam volumetric tomography. J Endod 2007;33:1121–32.

17. Nair MK, Nair UP. Digital and advanced imaging in endodontics: a review. J Endod2007;33:1– 6.

18. Nielsen RB, Alyassin AM, Peters DD, Carnes DL, Lancaster J. Microcomputed tomography:an advanced system for detailed endodontic research. J Endod 1995;21:561–8.

19. Patel S, Dawood A, Pitt Ford T, Whaites E. The potential applications of cone beamcomputed tomography in the management of endodontic problems. Int Endod J2007;40:818 – 823.

20. Yajima A, Otonari-Yamamoto M, Sano T, et al. Cone beam CT (CB Throne) applied todentomaxillofacial region. Bull Tokyo Dent Coll 2006;47:133– 41.

21. Scarfe WC, Farman AG, Sukovic P. Clinical applications of cone-beam computedtomograghy in dental practice. J Can Dent Assoc 2007;72:75– 80.

22. Trope M, Pettigrew J, Petras J, Barnett F, Tronstad L. Differentiation of radicularcyst and granulomas using computerized tomography. Endod Dent Traumatol1989;5:69 –72.

23. Simon JHS, Enciso R, Malfaz JM, Rogers R, Bailey-Perry M, Patel A. Differentialdiagnosis of large periapical lesions using cone-beam computed tomography mea-surements and biopsy. J Endod 2006;32:833–7.

TABLE 3. Results of Imaging Diagnostic Tests with CBCTPAI, on the Basis of Criteria Established from Measurements Corresponding to Periapical Radiolucency("0.5–1 mm to "8 mm), with 2 Additional Variables, Expansion of Cortical Bone (E) and Destruction of Cortical Bone (D) (n % 1014)

CBCT

n 0 1 2 E D 3 E D 4 E D 5 E D TotalPeriapical radiography 0 386 53 83 16 1 42 21 3 45 25 4 4 3 1 613

1 10 26 44 6 — 10 3 — 4 1 2 1 — — 952 — 3 51 14 1 58 33 5 44 22 8 4 2 1 160E — — — — — — — — — — — — — — —D — — — — — — — — — — — — — — —3 — — — — — 22 14 2 48 32 7 17 8 7 87E — — — — — — — — — — — — — — —D — — — — — — — — — — — — — — —4 — — 1 — — — — — 33 17 8 13 6 6 47E — — — — — — — — — — — — — — —D — — — — — — — — — — — — — — —5 — — — — — — — — — — — 12 6 5 12E — — — — — — — — — — — — — — —D — — — — — — — — — — — — — — —Total 1014

CBCTPAI, Cone Beam Computed Tomography Periapical Index.

Clinical Research


24. Laux M, Abbott PV, Pajarola G, Nair PNR. Apical inflammatory root resorption: acorrelative radiographic and histological assessment. Int Endod J 2000;33:483–93.

25. Landis JR, Koch GG. The measurement of observer agreement for categorical data.Biometrics 1977;33:159 –74.

26. Lofthag-Hansen S, Hummonen S, Gröndahl K, Gröndahl H-G. Limited cone-beam CTand intraoral radiography for the diagnosis of periapical pathology. Oral Surg OralMed Oral Pathol Oral Radiol Endod 2007;103:114 –9.

27. Nicopoulou-Karayianni K, Bragger U, Patrikiou A, Stassinakis A, Lang NP. Imageprocessing for enhanced observer agreement in the evaluation of periapical bonechanges. Int Endod J 2002;35:615–22.

28. Weiger R, Hitzler S, Hermle G, Löst C. Periapical status, quality of root canal fillingsand estimated endodontic treatment needs in an urban German population. EndodDent Traumatol 1997;13:69 –74.

29. Katsumata A, Hirukawa A, Noujeim M, et al. Image artifact in dental cone-beam CT.Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2006;101:652–7.

Clinical Research


cone beam pai estrela et al 2008

Documents