QUANTIFICATION OF INTRA OSSEOUS BONE DEFECTS BY CONVENTIONAL VERSUS 3DIMENSIONAL IMAGING TECHNIQUES-AN IN VIVO STUDY

* Dr. Anukriti Chhabra 1 , Dr. Ashita S. Uppoor 2 , Dr. Swati Pralhad 3 , Dr. Neelesh Singh 4 , Dr. Dilip G Nayak 5 . 1. BDS, Post graduate student, Department of Periodontology, Manipal College of dental sciences, Manipal University, Mangalore. 575001 2. MDS,Associate Dean and Professor, Department of Periodontology, Manipal College of dental sciences, Manipal University, Mangalore. 575001 3. MDS, Associate Professor, Department of Periodontology, Manipal College of dental sciences, Manipal University, Mangalore. 575001 4. MDS, Former Post graduate student, Department of Periodontology, Manipal College of dental sciences, Manipal University, Mangalore. 575001 5. MDS, Dean and Professor, Department of Periodontology, Manipal College of dental sciences, Manipal University, Mangalore. 575001

treatment modality and in determining prognosis. Current approaches to accurately diagnose periodontal disease include clinical probing of gingival tissues and an adjunctive use of radiographs to evaluate osseous structure. 2 Current conventional imaging includes bitewings, peri-apical (PA) and panoramicradiographs, with the main radiographic technique used for periodontal diagnosis being the paralleling extension cone method 3. These are easily acquired, cheap and provide high-resolution images; however they are not without limitations. The chief drawback is the problem of conspicuity which is largely the result of the presence of overlying anatomic structures and the representation of a 3D structure depicted by a two dimensional (2D) image. 4 The difficulty to reproduce angles over time leads to lack of standardisation. 5 There is ample research demonstrating that funnel-shaped or lingually located defects cannot be detected 6 and that destruction of the buccal plate can be undiagnosed or undistinguished from lingual defects 7. This makes it difficult to obtain an effective evaluation of periodontal bone loss, especially crater defects. Further studies comparing radiographs to pre-surgical measurements concluded that bone loss can be underestimated by; 1.5 mm, with large variations between examiners 8 . Consequently, traditional radiography remains a limited diagnostic tool 9. To address these issues, computed tomography (CT) has been explored because of its ability to produce accurate three-dimensional imaging without distortion. However, limitations such as high radiation, machine size, machine cost, complexity, and relatively low resolutionhave made this approach impractical 10 . Recently, cone beam computed tomography (CBCT) has turned this concept into potential reality because these lower-cost small machines produce high-quality data. The cone beam computed tomography has been used for imaging of the temporo-mandibular joint 11, evaluating oral osseous lesions 12, assessing maxillofacial deformities 13 and in preoperative planning of dental implants 14 . It may be adequate to visualize periodontal alveolar bone changes and emerge as a feasible tool in periodontal diagnosis. Yet there is little research to establish the use of this technique as a valid method for periodontal bone measurements.
Research comparing the use of CBCT to 2D radiographs in artificial bone defects has shown that CBCT has a sensitivity of 80-100% in the detection and classification of bone defects, while intraoral radiographs present a sensitivity of 63-67%. 15 Invitro studies involve no patient motion. Motion can, and often does, leads to CBCT image degradation. 16 In vivo research on this subject is also very limited to provide us with strong evidence. With these considerations in mind and with reference to current literature , the aim of this study was to verify the diagnostic accuracy of CBCT for intra-osseous periodontal bone defects, linearly, by using the peri-apical radiograph and direct intra-surgical measurement as a reference standard

Material and methods:-
The study was conducted in the Department of Periodontology, Manipal college of Dental Sciences, Manipal University, Mangalore, Karnataka. The study was initiated after approval from the Institutional Ethical committee. The study employed a prospective, double-blinded, comparative cross-sectional design.
Patients who were systemically healthy, vital teeth in whom interproximal intra-bony defects were present (>5mm), 2 or 3 walled defects and vertical bone defects of ≥2 mm on IOPA (intra oral peri-apical) radiograph were included in the study. Subjects who were immune-compromised, smokers, pregnant and lactating women, those taking any drugs (steroids, anticoagulants, anti-epileptics etc.) which are known to affect the healing and clotting mechanisms, causing gingival enlargement, and those with horizontal defects were excluded from the study.
The study protocol was in accordance with the ethical principles of the Declaration of Helsinki of 2008 as revised in 2013. Patients gave written informed consent to participate in the study. All the subjects signed a written informed consent form prior to being included in the study.
A total of five patients were selected. All patients who were undergoing non-surgical periodontal therapy with an additional need for surgical interventions were consequently selected from January 2014 to March 2014.
Clinical examination of all the subjects included measurement of oral hygiene status(Full mouth Plaque & Gingival index, pocket probing depth (PD) and clinical attachment loss (CAL).AUNC-15(GDC,PCPUNC-15) graduated periodontal probe calibrated to the nearest millimetre was used( Figure 1).The PD was recorded from the gingival margin to the base of the pocket. The CALwas determined by measuring the distance from the cement enamel junction (CEJ) to the base of the pocket. Upon clinical examination if the patient met the inclusion criteria, they were subjected to a routine peri-apical radiograph with grid using the Long cone paralleling technique. For a more detailed view a CBCT scan for the involved tooth was also taken. So, each patient was subjected to an intra-oralperiapical radiograph and a CBCT scan.
The IOPA radiograph was done for each selected site using a long cone paralleling technique where the film was placed parallel to the specific tooth with the central ray of the x-ray device perpendicular to the tooth of interest. The same film holder (XCP, RINN DENTSPLY) was used for all the radiographs .The F-speed film (Kodak Care stream Health Inc. New York, USA) mount with millimetre grid scale was used. All radiographs were digitalized using a digital camera (Cannon IXUS 105; 12.1 mega-pixels) and transferred to the computer in JPEG format. The IOPA radiograph of the area of interest was thoroughly examined prior to surgery. (Figure 2) CBCT scanning was performed using the clinically available machine (ProMax3DMid ProFacePLANMECA).The images were obtained using CBCT, and measurements were performed using the appropriate software.The present investigation used reconstructed 1.0 mm slices(voxel size)because this increment closely represents the same error of measurement when using a periodontal probe and because smaller slices decreases the image resolution. 17 Imageswere analysed by axial, cross-sectional and sagittal reconstructions The kVp , mA and scan duration was preset on the machine and inherently varied from patient to patient on the basis of built , jaw size and field of view. (Figure 3) All Peri-apical films were analysed using theImage J software, designed by National Institute of Health (NIH).The CBCT scanners inbuilt software had an electronic measuring tool to the nearest hundredth of a millimetre. All CBCT images were saved in DICOM format (Digital Imaging and Communications in Medicine) and were identical for each measurer as they were established by the same investigator, saved, and loaded at each session. The examiners used only one computer, and screen settings remained unchanged (1280 x1024 resolution). The examiners had the ability to change bone density and the size of the image to aid in optimal viewing. Grid markings on the radiographs and the CBCT scanned images were used to measure the following two measurements: (Figure 2 Periodontal bone defects were then measured during the surgical intervention using the cement-enamel junction, crest of the defect and base of the defect as the reference points. The direct surgical measurements were made using a UNC-15 graduated periodontal probe placed in a line parallel to the long axis of the tooth. The measurements taken were same as that recorded for CBCT and IOPA. Measurements falling between two marks of the periodontal probe were rounded up to the nearest millimetre. (Figure 4) Statistical Analysis:-Statistical analysis was performed using statistical software(SPSS.v.20). Coefficients of variation, using Pearson's method, were calculated to determine variabilitywithin the examiners and among the techniques. Interclass correlation was done to compare the accuracy of CBCT, IOPA and intra-surgical measurements. Paired t test was used to assess if the means of each of the six groups were statistically different from each other. The six groups are listed in the table.
Finally, A sample size of 6 achieved 83% power to detect a difference of -0.6 between the null hypothesis mean of 0.0 and the alternative hypothesis mean of 0.6 withanestimated standard deviation of 0.4 and with a significant level(alpha) of 0.5000 using a two-sided one-sample t-test.
Since 6 is a very small value for statistical analysis, the study had a sample size of 10.

Results:-
Five patients who met the inclusion criteria were selected for the study from January to March 2014. Atotal of ten defects were assessed and 60 measurements were performed. For measurement and validation purposes, both singleand multi-rooted teeth were considered. In order to represent different cortical and cancellous bone structures, both maxillary and mandibular teeth were included in the study. In order to classify the presence of alveolar bone loss, a distance of 3 mm from the CEJ to the Alveolar crest was used as the parameter of normality. The intra-surgical measurements are the gold standard parameter.
All defects were angular defects and the measurements were confined only to the inter-proximal area. The CBCT allowed for visualisation of defects on the buccal and lingual aspect as well. However the buccal and lingual defect measurements were not taken into consideration as these were masked in the IOPA.
The height (CEJ-C) and depth (CEJ-A) of each interproximal defect for each method was compared against the intra-surgical measurement.
Correlations between the two examiners varied between 0.09 and 0.99 The mean heights of the defects were6.15mm, 5.28mm and 5.3mm respectively for IOPA,CBCT and intra-surgical method. The IOPA measurements showed a statistically significant difference from the other two methods. There was no statistically significant difference with respect to CBCT versus intra-surgical measurements. (Table 1, figure  5) With respect to the defect depth, means were 9.3mm, 8.46mm and 8.28mm respectively for IOPA, CBCT and intrasurgical method. The results were statistically significant for IOPA versus the other two methods, but were not significant for CBCT versus intra-surgical. (Table 1, figure 5) The interclass correlation depicted thatthe IOPA measurements were only 74.3% accurate as compared to the standard intra-surgical whereas the CBCT measurements were 86.5 %. ( Figure 6       Interclass correlation between the three modalities depicting that the IOPA measurements were only 74.3% accurate as compared to the standard intra-surgical whereas the CBCT measurements were 86.5% accurate.

Discussion:-
The two-dimensional radiographic assessment and clinical examination were the only hopes for periodontal diagnosis till the advent of the three dimensional CT scan. The application of CT scan also deemed impractical due to size of the machine, radiation dose and cost. Overcoming all these limitations was the Cone Beam computed tomography, but research to substantiate its use for periodontal diagnosis has been limited. In the insight of literature, a systematic review reported that out of all the studies utilizing CBCT, only 3% were in the field of periodontology and most of them were in-vitro 18. CBCT facilitates to obtain a 360 degree image of the skull due to the generation of conic shaped beams as opposed to the fan shaped X-ray beams of the conventional CT scan machine. CBCT also results in a 15 fold reduction in the radiation absorbed by collimation of the primary X-Ray beam to a smaller area of interest 19,20. There are many studies that substantiate the fact that the radiation observed via CBCT is lesser than that by the CT scan machine. Honda et al 11 reported a reduction from 160 to 1.19 msV. Scaf et al reported a radiation exposure of 1.031 mSv(maxilla) & 2.426 mSv (mandible) using CT scan as opposed to 0.022 mSv in one study or to 0.15mSv for a full mouth series 21 . The NCRPM (National Council on Radiation Protection and Measurements) supplants the claim that the CBCT's radiation dose is within the conservative limits even at the highest CBCT exposure settings. It is likely that radiation exposure will further reduce as technology advances. 22 In spite of the CBCT being recommended as a dose sparing technique, the image quality eventually depends on the actual dose applied which varies slightly among different manufacturers and patient built. In this study we used a voxel size of 1mm as this closely represents the same error of measurement when using a periodontal probe and using smaller slices tends to decrease the image resolution 17  Feijó et al in their in vivo study used a smaller voxel size because a higher dose would provide higher-resolution images in an in vivo study. They compared the CBCT measurements only to intra-surgical measurements. Since we used the 2-D radiographs and the intra-surgical measurements for reference, a voxel size of 1mm was used. However, more research, including outcome assessment of various exposure and reformatting protocols and evaluation of the diagnostic validity during clinical follow-up, is required for proper justification of various CBCT applications in dental and maxillofacial radiology.
Out of all the in-vivo investigation on this subject, 3 were case reports. 23,24,25 One study by K de FariaVasconcelos et alcompared the CBCT measurements to the measurements of 2-D digital images retrospectively 15 and another prospective study by Feijó et al prospectively compared the CBCT measurements to intra-surgical for horizontal defects. 26 Our study was an in vivo study and the first study as such which compared all the three modalities i.e.IOPA, CBCT and intra-surgical measurements for intra-bony angular defects. This study closely represents the day to day clinical scenario with all the relevant standardisations taken into consideration.
Our study results showed that with respect to the depth of the defect, the IOPA showed a mean over estimation of 1.02 mm as compared to intra-surgical, and the CBCT showed a mean over estimation of 0.18mm.With respect to defect height, the IOPA showed a mean over estimation of 0.85mm as compared to intra-surgical, and the CBCT showed a mean underestimation of 0.02mm.Also the buccal and lingual extent of the defect were masked in the IOPA so the IOPA did not allow us to visualize deepest point of the defect if it lied in the vestibular areas. The CBCT images however allowed clear visualisation of the buccal and lingual extent of the defect. The exact morphology of the defect could thus be visualised. The IOPA still led to a mean over estimation of the defect measurements in the inter-proximal areas possibly because of errors in projection geometry and difference in the film placement for the different areas of the jaw. It was seen that there was more over-estimation in the anterior region, but the posterior region either over or underestimated.
Our study results were in agreement with an in vitro study by Vandenberghe et al who reported a significant difference between the 2-D radiographs and the CBCT. At 0.4mm voxel size, a mean error of 0.29 mm was detected on the 2D-CCD as compared to the CBCT, indicating that the CBCT demonstrated values closer to the gold standard 27 .Similarly;Misch et al showed that CBCT measurements were as accurate as direct measurements using a periodontal probe inexperimentally created buccal and lingual defects. 10 Feijó et al in their human study reported no statistically significant difference between clinical and CBCT measurements for horizontal bone defects in the maxillary posterior region. We are uncertain why the results of the present study differed from that of Feijo et al.It is possible that it can be due to the difference in the type of defects of both the studies. Another probable cause could be that we have taken all the areas of the jaw in to consideration whereas Feijo et al had a sample that consisted only of maxillary posterior teeth. The difference in the morphology and the result of difficult visualization during the clinical examination, mainly in the distal aspects may play a role as well.
It is worth noting that a 0.5-mm discrepancy between clinically and radio-graphically estimated bone levels is considered clinically acceptable. 5 The clinical measurements were obtained using probes with an accuracy of 1 mm, whereas CBCT measurements allowed an accuracy of up to three decimal places. These differences can cause some discrepancy in the comparison. So, further clinical studies are necessary to lay down certain guidelines and criteria's which will work as sure shot indications for the use of 3D imaging methods in periodontal diagnosis and treatment planning.

Conclusion:-
CBCT measurements compared well to the intra-surgical measurements in the quantification of intra-osseous periodontal bony defects. Considering the several advantages, limitations and risks of both modalities, though CBCT has higher cost and effective radiation dose, it provides more accurate details than IOPA especially in determining the size of the interproximal defects and the morphology of angular defects where the apex of the defect lies in the buccal or lingual area.One may conclude that the use of CBCT should only be reserved for relatively more complex periodontal treatment planning such as prognostic planning and surgery for complex periodontal defects. It should not be used explicitly for every case, however when in doubt we must not think twice before using the CBCT images as a cue.