QUINTESSENCE INTERNATIONAL PERIODONTOLOGY

Milena M. Cimbaljevic Clinical and CBCT-based diagnosis of furcation involvement in patients with severe periodontitis

Milena M. Cimbaljevic, DDS1/Rubens R. Spin-Neto, DDS, MSc, PhD2/Vesna J. Miletic, BDS, MSc, PhD3/Sasa M. Jankovic, DDS, MSc, PhD4/Zoran M. Aleksic, DDS, MSc, PhD4/Natasa S. Nikolic-Jakoba, DDS, MSc, PhD5

Objective: The aim of this study was to compare the use of ation. Agreement between the evaluation methods was pres- periodontal probing and cone beam computed tomography ent in 46.9% of cases (63.3% in maxilla, 45.0% in mandible). FI (CBCT) images in the diagnosis of furcation involvement (FI) in detected clinically was confirmed by means of CBCT in 24% of patients with chronic generalized severe periodontitis. the evaluated sites. The largest agreement (73.7%) in FI detec- Method and Materials: Fifteen patients with chronic gener- tion was found in the distopalatal maxillary sites between alized severe periodontitis were included in this study. In total, CBCT and clinical probing. The smallest agreement (36.6%) 174 furcation sites (all in molar teeth) were analyzed. FI was was found in the buccal sites of the mandibular molars, in assessed at three sites (buccal, mesiopalatal, and distopalatal) which 63.3% of FI were detected using CBCT only, but not clin- of maxillary molars, and at two sites (buccal and oral) of man- ically. Conclusion: The number of FI detected by means of dibular molars. FI was assessed both clinically (periodontal CBCT was larger than by means of periodontal probing. In probing) and on CBCT images, using a dichotomous scale those cases in which chronic generalized severe periodontitis is (present/absent). The agreement between clinical and CBCT- clinically diagnosed, and surgical treatment is necessary, CBCT based findings was calculated. Results: FI were more often may be suggested as an adjunct tool for FI assessment. detected by means of CBCT than by means of clinical examin- (Quintessence Int 2015;46:863–870; doi: 10.3290/j.qi.a34702)

Key words: cone beam computed tomography, furcation involvement, periodontal disease

Periodontitis, a biofilm-induced destructive disease of olar bone leads to bone defects around the teeth and dental supporting tissues, is the most frequent cause in the interradicular (furcation) region.2 The prognosis associated to molar teeth loss.1 The gradual loss of alve- of the disease and the success of therapy depends on an accurate assessment of the residual alveolar bone, 2 1 PhD student, Department of Periodontology, School of Dental Medicine, Univer- particularly in the interradicular regions. Teeth that sity of Belgrade, Belgrade, Serbia. have lost attachment to the level of the furcation are 2 Assistant Professor, Department of Dentistry, Section of Oral Radiology, Aarhus University, Aarhus, Denmark. considered to have a furcation involvement (FI). FI vary 3 Assistant Professor, Department of Restorative Dentistry and Endodontics, in horizontal and vertical depth as a result of features School of Dental Medicine, University of Belgrade, Belgrade, Serbia. such as cervical enamel pearls, root trunk length, furca- 4 Associate Professor, Department of Periodontology, School of Dental Medicine, University of Belgrade, Belgrade, Serbia. tion entrance dimensions, root anatomy, and variations 2 5 Assistant Professor, Department of Periodontology, School of Dental Medicine, in the furcation root anatomy. Periodontal diagnosis of University of Belgrade, Belgrade, Serbia. a FI is usually based on the clinical examination, Correspondence: Professor Natasa Nikolic-Jakoba, Department of Peri- assisted by dental radiographs.3 odontology, School of Dental Medicine, University of Belgrade, Dr Subotica 8, Belgrade, Serbia. Email: [email protected]

VOLUME 46 • NUMBER 10 • NOVEMBER/DECEMBER 2015 863 QUINTESSENCE INTERNATIONAL Cimbaljevic et al

major advantage of CBCT the fact that a much lower Table 1 Characteristics of the study population patient dose is achieved than in conventional CT, con-

Characteristic Details sidering a volume of the same size. On the other hand, Sex (males/females) 6/11 the major drawback associated to CBCT would be the Age (years; mean ± SD) 44.53 ± 8.37 increased effect of scatter radiation on image quality, Probing depth (mm; mean ± SD) 4.81 ± 0.98 leading to artefacts.3 Besides that, the radiation dose Clinical attachment level (mm; mean ± SD) 5.12 ± 1.05 given to the patient is higher than when other 2D meth- ods are used.12 There is little evidence in the literature regarding the comparison of clinical evaluation (prob- During the clinical examination, the degree of FI can ing) and 3D imaging methods on the assessment of FI. be assessed by probing using a Nabers probe. Based on In two studies, CT findings were compared to the clin- these findings, Hamp et al4 suggested a classification ical diagnosis.13,14 Darby et al15 reported the correlation system of FI according to the horizontal loss of peri- between the data found in the patient’s records and the odontal tissues in the furcation area, graded in three CBCT findings (retrospective study). The agreement degrees. However, probing is often related to the depth between presurgical clinical evaluation and 3D imaging of probe penetration into the inflamed connective tis- methods on the assessment of FI remains unclear. sue, and not on the actual depth of the interradicular The aim of the present study was to compare the defect.5 Additionally, the use of the currently accepted use of periodontal probing and CBCT images in the clinical detection and classification system may be diagnosis of FI in patients with chronic generalized affected depending on the root morphology, the con- severe periodontitis. figuration of the residual inter- and peri-radicular bone, the length of the root trunk, and the degree of root METHOD AND MATERIALS separation. These parameters might influence the diag- nosis and therefore the definition of an appropriate Fifteen patients (four men and 11 women, aged 35 to treatment plan.6 60 years; mean age 44.5 ± 8.4 years) suffering from The importance of radiographic images in assessing chronic generalized severe periodontitis16 participated periodontal disease progression has been previously in the study (Table 1). Briefly, chronic generalized described.3 Two-dimensional (2D) imaging modalities, severe periodontitis was defined on the basis of extent such as digital periapical images, are most frequently and severity, where > 30% of sites of all present teeth used.3 Although useful, these images are not accurate were affected, with the attachment and bone enough regarding the assessment of FI, due to the loss > 5 mm.16 overlap of anatomical structures and the lack of Patients were consecutively recruited from May three-dimensional (3D) information.3 The accuracy of 2013 to June 2014 from the pool of patients at the 2D radiographic methods in FI assessment increases Department of Periodontology, School of Dental Medi- with the degree of attachment loss.7 To allow an earlier cine, University of Belgrade, Serbia. The study was detection of FI, a more accurate assessment method is approved by the Ethics Research Committee of the needed. School of Dental Medicine, University of Belgrade, Ser- The use of cone beam computed tomography bia (ethics approval no. 36/2). All patients were thor- (CBCT) devices in dentistry has expanded, allowing the oughly informed about the purpose of the study and 3D visualization of dental and maxillofacial anatomical gave their written consent before clinical examination. structures in an easier manner.8 At the same time, only a The inclusion criteria were the presence of at least two few in-vivo studies on the use of CBCT for FI assessment intrabony defects, with probing depth ≥ 6 mm in both are present in the literature.6,9-11 Overall, they cite as the jaws, indicated for periodontal surgery. In addition,

864 VOLUME 46 • NUMBER 10 • NOVEMBER/DECEMBER 2015 QUINTESSENCE INTERNATIONAL Cimbaljevic et al

ab

Figs 1a to 1d Intraoral clinical photo- graph (a) and CBCT with sagittal, axial, and coronal sections (b to d) of the maxillary left first molar. Buccal furcation involve- ment of the maxillary left first molar was qualified as “present” on CBCT scans, but was not detected clinically by probing (probing depth = 1 mm, clinical attach- ment level = 1 mm). cd

patients should have no systemic diseases, and not be inserting the probe horizontally into the furcation to lactating or pregnant. the deepest point of the defect (Fig 1a). Clinical exam- Nonsurgical periodontal treatment (ie, scaling and inations were performed by two trained periodontists root planing) performed by two well-experienced clin- (NNJ and SJ), who were previously calibrated (Cohen’s icians, was performed at least 6 weeks before inclusion ĸ = 0.697). The examinations were based on a dichoto- in the study on all teeth with periodontal probing mous scale (present/absent). Both observers examined depths of ≥ 4 mm. The period of 6 weeks was chosen the patients in the same clinical session, but separately. since at that time further treatment plan and deci- The examination was performed by each examiner sion-making related to the need for periodontal sur- without the presence of another, with a 30-minute gery was made according to the protocol used at the interval between examinations, and examiners were school.16,17 After the completion of the nonsurgical blinded to the findings of each other. periodontal treatment, clinical evaluation of FI was Patients were further scanned using CBCT, which performed, using a Nabers probe (PQ2N, Hu-Friedy). A would be used in the planning of the scheduled peri- total of 38 maxillary and 30 mandibular molars were odontal surgery for the treatment of the intrabony assessed (total of 174 furcation sites). Third molars were defects. For that, patients were examined using a not evaluated. FI was observed at three sites (buccal, high-resolution CBCT unit (SCANORA 3Dx, Soredex). mesiopalatal, and distopalatal) of maxillary molars, and Examinations were performed using a 80 × 100 mm at two sites (buccal and oral) of mandibular molars, by field of view, 0.25 mm voxel size, 90 kV tube voltage, 10

VOLUME 46 • NUMBER 10 • NOVEMBER/DECEMBER 2015 865 QUINTESSENCE INTERNATIONAL Cimbaljevic et al

Table 2 Sample distribution (number of teeth, number of evaluated sites, and percentage)

Maxilla Mandible Teeth Sites % Teeth Sites % First molar 22 66 58 12 24 40 Second molar 16 48 42 18 36 60 Total 38 114 100 30 60 100

Table 3 Number (and percentage) of FI detected clinically and using CBCT, and agreement between the methods, according to the jaw (maxilla/mandible)

Clinical assessment Maxilla Mandible Total CBCT assessment Absent Present Absent Present Absent Present Absent 25 (22.9%) 0 (0.0%) 15 (25.0%) 0 (0.0%) 40 (22.9%) 0 (0.0%) Present 42 (36.8%) 47 (41.3%) 33 (55.0%) 12 (20.0%) 75 (43.1%) 59 (24.0%) Total 114 (100.0%) 60 (100.0%) 174 (100.0%) Agreement 63.3% 45.0% 46.9%

mA tube current, and 2.4 seconds active scanning time. molar, or buccal and oral for mandibular molars). The CBCT volumes in the DICOM format were exported Agreement was declared when both the clinical and to dedicated software (OnDemand3D, Cybermed). One the CBCT assessments were equal. trained observer (MC), blinded to the clinical findings, assessed the full CBCT volumes for FI, also based on a RESULTS dichotomous scale (present/absent) (Figs 1b to 1d). CBCT assessments were performed on two separate Sample distribution is presented in Table 2. The number occasions, within a 4-week interval. Intraobserver and percentage of FI diagnosed clinically and using agreement was 0.94. For the assessment, the same CBCT is presented in Table 3, considering all examined software was used. The volumes were assessed using a regions. As an overall result, FI were more often detected 17-inch monitor (VA2231WA-LED, ViewSonic), with using CBCT than by means of clinical examination. 1,280 × 1,024 resolution. Section thickness was set to Agreement between the evaluation methods was pres- 1 mm in all sectioning planes. ent in 46.9% of cases. All disagreements were related to All collected data were organized and evaluated a FI diagnosed in CBCT but not clinically (probing). using dedicated software (SPSS 13.0, IBM). Results were calculated and presented as an agreement between Agreement between the clinical findings the two evaluation methods (clinical and CBCT), and CBCT data in the maxilla according to the jaw (maxilla or mandible) and site Diagnosis for maxillary sites presented a stronger (buccal, mesiopalatal, and distopalatal for maxillary agreement than for those in the mandible (63.3% and

866 VOLUME 46 • NUMBER 10 • NOVEMBER/DECEMBER 2015 QUINTESSENCE INTERNATIONAL Cimbaljevic et al

Table 4 Number (and percentage) of FI detected clinically and using CBCT, and agreement between the methods, according to the site (buccal, mesiopalatal, distopalatal), in maxillary molars

Clinical assessment Buccal Mesiopalatal Distopalatal CBCT assessment Absent Present Absent Present Absent Present Absent 9 (23.7%) 0 (0.0%) 9 (23.7%) 0 (0.0%) 7 (18.4%) 0 (0.0%) Present 15 (39.5%) 14 (36.8%) 17 (44.7%) 12 (31.6%) 10 (26.3%) 21 (55.3%) Total 38 (100.0%) 38 (100.0%) 38 (100.0%) Agreement 60.5% 55.3% 73.7%

Table 5 Number (percentage) of FI detected clinically and using CBCT, and agreement between the methods, according to the site (buccal, oral), in mandibular molars

Clinical assessment Buccal Oral CBCT assessment Absent Present Absent Present Absent 7 (23.3%) 0 (0.0%) 8 (26.7%) 0 (0.0%) Present 19 (63.3%) 4 (13.3%) 14 (46.6%) 8 (26.7%) Total 30 (100.0%) 30 (100.0%) Agreement 36.6% 53.4%

45%, respectively). The strongest agreement between these regions, in 46.6% of cases, FI were more often CBCT and clinical evaluation for the detection of FI was detected in CBCT scans than clinically. found in the distopalatal maxillary site (agreement of 73.7%, Table 4). The difference between the diagnostic DISCUSSION methods was less prominent in the buccal sites (agree- ment in 60.5% of cases), while the mesiopalatal sites Literature is yet limited regarding CBCT-based assess- showed the highest disagreement in the maxilla ment of FI in patients. Therefore, in the present study, (agreement in 55.3% of cases). In aforementioned sites the existence of furcation involvement (FI) in the maxil- FI were detected in CBCT but not clinically in 26.3%, lary and mandibular molars, comparing clinical examin- 39.5%, and 44.7% of cases, respectively. ation (based on probing), and radiographic assessment (based on CBCT) was investigated. CBCT more fre- Agreement between the clinical findings quently revealed FI in the examined sites, compared to and CBCT data in the mandible probing. Considering both assessment methods, 22.9% The weakest agreement (36.6%) was found in the buc- of the sites presented no FI. Regarding those sites in cal sites of the mandibular molars (Table 5). For these which at least one assessment method indicated the sites 63.3% of FI were assessed using CBCT only, but presence of FI, clinical detection (probing) was con- were not detected clinically. The difference between firmed using CBCT in 24.0% of the cases, while in 43.1% the diagnostic methods was less prominent in the oral probing underestimated the CBCT findings. mandibular sites (agreement in 53.4% of cases). In

VOLUME 46 • NUMBER 10 • NOVEMBER/DECEMBER 2015 867 QUINTESSENCE INTERNATIONAL Cimbaljevic et al

Interestingly, the biggest difference between prob- assessment of FI was investigated in several ex-vivo ing and CBCT findings was seen in the buccal region of studies. CBCT has proven to be highly accurate in the mandibular molars, which are also considered to be detection of initial FI simulated on pigs’ mandibles.22 the most accessible regions for probing. It has been The accuracy of CBCT for the assessment of FI was com- shown that the ideal probe position and good accessi- pared with that of direct probing on a dry human skull, bility during the clinical evaluation of FI do not guaran- having the clinical truth as the gold standard, and tee FI detection.18 Further, double measurements showed that none of the used techniques led to false during clinical examination are associated with higher positives or false negatives for FI detection for that spe- variability in molars than in single-rooted teeth.19 cific study design.23 Using the same study design, Van- In the present study, the clinical examination of FI denberghe et al24 compared the accuracy of CBCT was performed 6 weeks after nonsurgical periodontal images with digital periapical radiographs in the detec- therapy. At that point periodontal wound healing was tion of FI, and showed that FI was correctly detected in most likely completed, and discrete furcal attachment all CBCT images, but in only half of the intraoral images. loss was masked mostly by the presence of long epithe- Walter et al9 showed that CBCT provides more lial attachment. The omission of initial FI may be due to detailed information about the degree of FI compared the fact that the furcation entrance was not properly to the clinical findings and periapical radiographs. Fur- located.18 The diameter of the furcation entrance has a ther, Walter et al10 indicated that CBCT evaluation of FI big impact on the accuracy in diagnostics of FI. Bower20 in maxillary molars was in strong agreement with intra- showed that 81% of furcation entrances surgical findings, which are considered to be the “gold were < 1.0 mm, and 58% of were < 0.75 mm. This ana- standard” in FI assessment. Namely, in that study CBCT tomical feature may be one plausible explanation for underestimated 14.7% of the cases and overestimated the highest mismatch at buccal furcation entrances of 1.3%, compared with the intrasurgical assessments.10 mandibular molars (63.3%). Positive correlation Qiao et al6 reported similar results in their study, in between the presence of cervical enamel projection which CBCT data under- or overestimated intrasurgical (CEP) and FI has previously been described. Mandibular findings in 17.6% of the cases. In all three aforemen- first molars showed the highest prevalence of CEP.21 It tioned studies, agreement between clinical (intrasurgi- should be taken into consideration that this anatomical cal) and CBCT findings was approximately 20%. This is characteristic may contribute to the highest mismatch compatible with the level of agreement found in the between clinical and CBCT findings in buccal furcation present study, when probing and CBCT findings were entrances of mandibular molars. Stronger agreement compared. It is important to keep these findings in between the examination methods was detected at mind, since they point to the fact that there could be oral furcation entrances of mandibular molars (53.4%). cases of “over-diagnosis” related to CBCT-based FI Surprisingly, the strongest agreement between the two detection, which would lead to “over-treatment”. Con- methods was found at distopalatal furcation sites of the trary, underestimation of FI could result in lack of maxillary molars, although that region is not clinically appropriate therapy, which could lead to the progres- easily accessible. sion of periodontal disease. Accurate detection of FI is a major challenge even A drawback of all evaluation methods based on for experienced clinicians. Proper diagnosis will lead to radiographic or tomographic images is patient’s expo- the correct treatment plan, which will then not be sure to radiation. As Low as Reasonable Achievable changed during the course of periodontal treatment. (ALARA) as a fundamental principle for diagnostic CBCT has been evaluated as a diagnostic tool in peri- radiology must be followed.25 In the present study, odontology because of its ability to display bone CBCT scanning was primarily used to assess periodontal defects in a 3D manner.6,9 The accuracy of CBCT in the (intrabony) defects in both jaws. All patients included in

868 VOLUME 46 • NUMBER 10 • NOVEMBER/DECEMBER 2015 QUINTESSENCE INTERNATIONAL Cimbaljevic et al

the study suffered from chronic generalized severe CONCLUSION periodontitis, and therefore, a field-of-view (FOV) of 80 × 100 mm was used, allowing that one CBCT acqui- The number of FI detected by means of CBCT was sition would show all sites. In this way, FI evaluation larger than by means of clinical probing. In those cases was performed as an additional diagnostic procedure. in which chronic generalized severe periodontitis is For FI evaluation, a smaller FOV (50 × 50 mm) should be clinically diagnosed, and surgical treatment is neces- used, which would result in dose reduction.26 One sary, CBCT may be suggested as an adjunct tool for FI could speculate on how much information CBCT assessment. images add to clinical probing in cases of FI involve- ment. Zappa et al27 stated that clinical FI diagnosis is ACKNOWLEDGMENTS limited, and should be supplemented by other diag- nostic methods. Data from CBCT could be helpful in This work was supported by the Ministry of Education and Science of preoperative evaluation of size and architecture of the the Republic of Serbia (Grant #41008). The authors thank Jörg K.F. Mudrak, DDS, DMD, Private Clinic, Ludwigsau, Germany, for his tech- defect and in surgical treatment plan optimization. nical assistance and help regarding computer-based analysis. Without precise information of FI, periodontal surgery may reveal unexpected findings, leading to changes in the treatment plan during the surgery,28 and causing REFERENCES unanticipated treatment costs (financial and tempo- 1. McFall WT Jr. Tooth loss in 100 treated patients with periodontal disease. A long-term study. J Periodontol 1982;53:539–549. 29 ral). This means that for those patients with chronic 2. Matthews DC, Tabesh M. Detection of localized tooth-related factors that generalized severe periodontitis a CBCT examination predispose to periodontal infections. Periodontol 2000 2004;34:136–150. prior to definitive treatment plan should be performed, 3. Mol A. Imaging methods in periodontology. Periodontol 2000 2004;34:34–48. 4. Hamp SE, Nyman S, Lindhe J. Periodontal treatment of multirooted teeth. in those cases in which surgical treatment is necessary. Results after 5 years. J Clin Periodontol 1975;2:126–135. Preliminary to the decision of having a CBCT examin- 5. Moriarty JD, Hutchens LH Jr, Scheitler LE. Histological evaluation of periodon- tal probe penetration in untreated facial molar furcations. J Clin Periodontol ation done, the clinician should consider the amount of 1989;16:21–26. additional information that would be provided, and in 6. Qiao J, Wang S, Duan J, et al. The accuracy of cone-beam computed tomog- raphy in assessing maxillary molar furcation involvement. J Clin Periodontol which extent the treatment plan would be altered. 2014;41:269–274. In the European Commission’s evidence-based 7. Hardekopf JD, Dunlap RM, Ahl DR, Pellau GB Jr. The “furcation arrow”. A reli- able radiographic image? J Periodontol 1987;58:258–261. guidelines for CBCT use it is written that: “in terms of 8. Mozzo P, Procacci C, Tacconi A, Tinazzi Martini P, Bergamo Andreis IA. A new detection of periodontal bone loss, laboratory studies volumetric CT machine for dental imaging based on the cone-beam tech- nique: preliminary results. Eur Radiol 1998;8:1558–1564. do not permit a comparison of CBCT with the primary 9. Walter C, Kaner D, Berndt DC, Weiger R, Zitzmann NU. Three-dimensional diagnostic method (ie, probing of pockets)”.12 Further- imaging as a pre-operative tool in decision making for furcation surgery. J Clin Periodontol 2009;36:250–257. more, the guidelines state that “the impact of three-di- 10. Walter C, Weiger R, Zitzmann, NU. Accuracy of three-dimensional imaging in mensional images upon management decisions and assessing maxillary molar furcation involvement. J Clin Periodontol 2010;37:436–441. treatment impact in clinical practice has not been con- 11. Walter C, Dagassan-Berndt DC, Kühl S, Weiger R, Lang NP, Zitzmann NU. Is sidered”.12 The guidelines suggest as a general direc- furcation involvement in maxillary molars a predictor for subsequent bone augmentation prior to implant placement? A pilot study. Clin Oral Implants tion that CBCT may “have a role to play in the manage- Res 2013;24:1–7. ment of complex periodontal defects for which surgery 12. SEDENTEXCT Project. Radiation protection n° 172: Cone Beam CT for dental and maxillofacial radiology. Luxembourg: European Commission Director- is the treatment option.”12 Despite the fact that there is ate-General for Energy, 2012. rapidly accruing literature on CBCT, there is still no cur- 13. Laky M, Majdalani S, Kapferer I, et al. Periodontal probing of dental furcations compared with diagnosis by low-dose computed tomography: a case series. rent evidence-based guideline on its necessity and use J Periodontol 2013;84:1740–1746. for periodontal treatment planning and implementa- 14. Pistorius A, Patrosio C, Willershausen B, Mildenberger P, Rippen G. Periodontal probing in comparison to diagnosis by CT-scan. Int Dent J 2001;51:339–347. tion, including the handling of FI.

VOLUME 46 • NUMBER 10 • NOVEMBER/DECEMBER 2015 869 QUINTESSENCE INTERNATIONAL Cimbaljevic et al

15. Darby I, Sanelli M, Shan S, et al. Comparison of clinical and cone beam com- 23. Fleiner J, Hannig C, Schulze D, Stricker A, Jacobs R. Digital method for quanti- puted tomography measurements to diagnose furcation involvement [Epub fi cation of circumferential periodontal bone level using cone beam CT. Clin ahead of print 16 Dec 2014]. Int J Dent Hyg, doi: 10.1111/idh.12116. Oral Invest 2013;17:389–396. 16. Armitage GC. Periodontal diagnoses and classifi cation of periodontal diseas- 24. Vandenberghe B, Jacobs R, Yang J. Detection of periodontal bone loss using es. Periodontol 2000 2004;34:9–21. digital intraoral and cone beam computed tomography images: An in vitro 17. Takei HH, Carranza FA. The surgical phase of therapy. In: Newman MG, Takei assessment of bony and/or infrabony defects. Dentomaxillofac Radiol HH, Klokkevold PP, Carranza AF (eds). Carranza’s Clinical Periodontology, 10th 2008;37:252–260. edition. St. Louis: Saunders 2006:881–886. 25. Farman AG. ALARA still applies. Oral Surg Oral Med Oral Pathol Oral Radiol 18. Eickholz P. Reproducibility and validity of furcation measurements as related Endod 2005;100:395–397. to class of furcation invasion. J Periodontol 1995;66:984–989. 26. Ludlow JB, Davies-Ludlow LE, Brooks SL. Dosimetry of two extraoral direct 19. Becherer CF, Rateitschak KH, Hefti AF. Vergleichende Sondierung mit ein digital imaging devices: NewTom cone beam CT and Orthophos Plus DS erelektronischen und einer manuellen Parodontalsonde. Schweiz Monatsschr panoramic unit. Dentomaxillofac Radiol 2003;32:229–234. Zahnmed 1993;103:715–721. 27. Zappa U, Grosso L, Simona C, Graf H, Case D. Clinical furcation diagnoses and 20. Bower RC. Furcation morphology relative to periodontal treatment. Furcation interradicular bone defects. J Periodontol 1993;64:219–227. root surface anatomy. J Periodontol 1979;50:366–374. 28. Walter C, Weiger R, Zitzmann NU. Periodontal surgery in furcation-involved 21. Hou G-L, Tsai C-C. Relationship between furcation involvement and molar maxillary molars revisited: an introduction of guidelines for comprehensive cervical enamel projections. J Periodontol 1987;58:715–721. treatment. Clin Oral Invest 2011;15:9–20. 22. Umetsubo OS, Gaia BF, Costa FF, Cavalcanti MGP. Detection of simulated 29. Walter C, Weiger R, Dietrich T, Lang NP, Zitzmann NU. Does three-dimension- incipient furcation involvement by CBCT: an in vitro study using pig mandi- al imaging off er a fi nancial benefi t for treating maxillary molars with furcation bles. Braz Oral Res 2012;26:341–347. involvement? A pilot clinical case series. Clin Oral Implants Res 2012;23: 351–358.