A controlled study on the diagnostic accuracy of panoramic and peri-apical radiography for detecting furcation involvement

doi:10.21203/rs.3.rs-41590/v2

Download PDF

Research article

A controlled study on the diagnostic accuracy of panoramic and peri-apical radiography for detecting furcation involvement

https://doi.org/10.21203/rs.3.rs-41590/v2

This work is licensed under a CC BY 4.0 License

Journal Publication

published 12 Mar, 2021

Read the published version in BMC Oral Health →

You are reading this latest preprint version

Background: The aim of this study was to determine diagnostic accuracy, sensitivity and specificity of panoramic and peri-apical radiography for detecting furcation involvement (1) and to evaluate the possible impact of clinical experience on these diagnostic parameters (2).

Methods: Periodontitis patients in need of an implant were retrospectively selected. Inclusion criteria were the presence of a CBCT, panoramic and peri-apical radiograph of the site of interest within a one-year time frame. All furcation sites were classified on the basis of CBCT using Hamp’s index (1975). These data were considered gold standard. Ten experienced examiners and 10 trainees were asked to assess furcation involvement for the same defects on the basis of corresponding panoramic and peri-apical radiographs. Absolute agreement, Cohen’s weighted kappa, sensitivity and specificity were calculated. In addition, ROC-curves were constructed.

Results: The study sample included 60 furcation sites in 29 multi-rooted teeth from 17 patients (10 females; mean age 62). On average 20/60 furcations were correctly classified on the basis of panoramic radiography, corresponding to weighted kappa of 0.209 indicative of slight agreement. Respective data for peri-apical radiography were 19/60 and 0.221. When recategorizing FI Grades into ‘no to limited FI’ (FI Grade 0 and I) and ‘advanced FI’ (FI Grade II and III), panoramic and peri-apical radiography showed low sensitivity (0.558 and 0.441, respectively), yet high specificity (0.791 and 0.790, respectively) for identifying advanced FI. Both showed diagnostic value given the area under the ROC-curve amounting to 0.79 and 0.69 for panoramic and peri-apical radiography, respectively. There was no significant difference between experienced periodontists and trainees (P = 0.257 versus P = 0.880).

Conclusion: Panoramic and peri-apical radiography are relevant in the diagnosis of FI given high specificity. These are best combined with furcation probing showing high sensitivity. Clinical experience does not seem to improve the accuracy of a radiological diagnosis of furcation sites.

Trial Registration: Patients were retrospectively registered.

Dentistry

Diagnosis

furcation involvement

periodontal disease

CBCT

panoramic radiography

intra-oral radiography

The diagnosis of periodontitis is based on clinical as well as radiographic examination(1)(2). Two-dimensional radiographic examination by means of peri-apical radiography, is still the standard method for assessing marginal bone loss(3). In addition, decay, root morphology and resorptions can be identified(4). Peri-apical radiographs are preferably taken by using the paralleling technique for good diagnostic quality(5).

Panoramic radiography, occasionally combined with some peri-apical radiographs, can be used as an alternative to a full-mouth series of peri-apical radiographs given a lower radiation dose (6). Among general dentists, there is a notable variation in the selection and use of radiographic methods in the assessment of periodontal diseases(7).

A detailed diagnosis of periodontitis is more challenging in the posterior dentition because of furcation and root proximities. In case of periodontal attachment loss around multi-rooted teeth, bone will disappear stepwise and the furcation may become involved(8). Such furcation involvement (FI) is a clinical parameter and determines the severity of pathological resorption of the supporting alveolar bone within a furcation(9). Interradicular osseous defects are associated with loss of tooth support. Furthermore, there is an association that ecological niches are situated in these interradicular defects. Ecological niches could be locus-specific risk factors(10) and induce further progression of pathological bone breakdown by means of apical downgrowth and spread of subgingival plaque between the root cones(4). First a widening of the periodontal space is seen in combination with inflammatory and cellular fluid exudation followed by epithelial proliferation into the access of the furcation(11).

The degree of attachment loss in the furcation of multi-rooted teeth is subdivided in gradations based on horizontal measurements as described by Hamp et al.(12). Grade 0 implies that the furcation is not accessible with a curved Nabers furcation probe. Grade Ι equals horizontal attachment loss up to 3mm. Grade ΙΙ means horizontal attachment loss exceeding 3mm, but no detectable “through and through destruction”. Grade III is horizontal “through-and-through” destruction of the periodontal tissues in the furcation. Vertical defects associated with intrabony pockets could also been seen as pattern of bone destruction(11). Given the presence of only 2 roots the furcation anatomy is rather easy in the mandible with only an access at the buccal and lingual aspect. In contrast, 3 roots form the furcation in the maxilla making the anatomy much more complex. As a result, 3 furcation areas can be identified around maxillary molars: buccal, mesiopalatal and distopalatal. In addition, variations exist in the morphology of multirooted teeth. For example, the length of the common root trunk and degree of separation of root cones. These variations are important in defining at what stage in the periodontal breakdown progress a furcation will become present(4).

In treatment planning, FI is an important criterion to attribute a prognosis to a tooth being possibly secure, doubtful or irrational to treat. This is based on the fact that FI is considered a locus minoris resistentiae increasing the risk for tooth loss. In this respect, McGuire & Nunn(13) described that a tooth with Grade III FI has an increased risk of becoming lost than a tooth with Grade II, respectively Grade I or no FI. Multirooted teeth with any FI are more at risk for further attachment loss hereby compromising the long-term prognosis(14). Given this, it is logic that FI should be considered in the decision-making process together with other dental- and patient-related factors. Furthermore, the reduced accessibility and complicated anatomical architecture of furcation defects limits the efficacy of initial periodontal therapy(15)(16). When residual pockets ≥ 6mm and full-mouth bleeding score of < than 30% persist, periodontal surgery might additionally be necessary(17).

Cone beam computed tomography (CBCT) is a relatively new 3D approach in oral imaging (8)(18). The images are captured by means of a pyramid-shaped beam. CBCT provides image reconstructions of dental anatomy and pathology in axial, coronal and sagittal planes making a detailed and accurate diagnosis possible(19). The degree of FI can be easily retrieved as CBCT is a digital three-dimensional reconstruction of a complex clinical situation, whereas panoramic and peri-apical radiographs provide two-dimensional images(8). In general, CBCT requires smaller radiation dose compared to conventional medical CT(20). For medium field of view (FOV) the radiation dose ranges from 9 to 560μSv. The effective radiation dose of an average panoramic image is 3-24.3μSv and amounts to 34.9-104.71μSv for an intra-oral full-mouth series (21)(22).

Two-dimensional radiographic examination is limited in assessing marginal bone levels and incipient inter-radicular bone loss because of overlap of bone and surrounding anatomic structures(8). Underestimation of alveolar bone loss ranges from 13-32% in panoramic radiography versus 9-21% in peri-apical radiography(23). Peri-apical radiography is also more accurate in detecting and assessing the dimensions of periodontal bone defects(24). However, detailed information on the accuracy of panoramic and peri-apical radiography for the detection of FI has not been published. Hence, the primary objective of this controlled study was to determine diagnostic accuracy, sensitivity (SENS) and specificity (SPEC) of panoramic and peri-apical radiography for detecting furcation involvement (1) and to evaluate the possible impact of clinical experience on these diagnostic parameters (2).

Patient and site selection

Patient demographics and radiographs were retrospectively selected from Ghent University Hospital records of patients who had been treated at the Department of Periodontology and Oral Implantology. Patients demonstrated varying severity and extent of periodontal disease, had at least one molar in situ without merged roots and had one or more implants placed following periodontal therapy. Given that, CBCTs were available from all patients. Additional inclusion criteria were the presence of a panoramic and peri-apical radiograph of the site of interest within a one-year time frame. The study was conducted in accordance with the Helsinki declaration of 1975 as revised in 2000. The protocol was approved by the ethical committee of the Ghent University Hospital (B670201523577).

Examiners

Ten staff members who had been working for at least 5 years as a periodontist and 10 postgraduate trainees in Periodontology and Oral Implantology participated in the evaluation of radiographs. As it was our intention to assess the diagnostic value of panoramic and peri-apical radiograph for assessing advanced FI in daily practice, clinicians were deliberately not trained and calibrated beforehand.

Assessment of FI on the basis of Cone Beam Computed Tomography (CBCT), panoramic and peri-apical radiographs

For every furcation site the degree of horizontal alveolar bone loss was established on CBCT (Planmeca ProMax® 3D Max, Helsinki, Finland) by two experienced clinicians (VC, MD). CBCTs were taken with a standardized protocol for three-dimensional implant planning (⌀100x55mm or ⌀100x90mm medium FOV with 200µm voxel size, 90kV tube voltage and 12 seconds scanning time). Mean mA was 5.6 and ranged between 4.5 and 7.1 depending on the skull size and weight of the patient. All CBCT volumes were available in DICOM format, analysed with Planmeca Romexis© software (Romexis 4.5.2.R, Helsinki, Finland) and carefully assessed in all different planes on a 24-inch monitor (Barco Eonis® MDRC-2224 BL, Kortrijk, Belgium) with dimmed surrounded light. The data on horizontal alveolar bone loss were transformed to the Hamp et al.(12) classification (= gold standard). Whenever a different Grade was found among the two experienced clinicians, the final Grade was determined following discussion.

Every clinician scored the same sites using the Hamp et al.(12) classification on the basis of panoramic (Planmeca ProMax dimax4 2DScara3 + Pan + Cephalostat GUI version 3.7.1.0.r, Helsinki, Finland) and peri-apical radiographs (Dürr Dental Vistascan Phosphor plates size 2+, Bietichheim-Bissingen, Germany) under optimal conditions: separately and in a quiet semi-dark room with dimmed surrounded light. A washout period of two months was respected to eliminate possible effects from previous measurements.

Panoramic images had been taken with 66kV tube voltage and 8mA. Digital peri-apical radiographs had been taken with Rinn XCP^® Paralleling holders (Dentsply^®, Weybridge, UK). The active exposure time ranged between 0.08 and 0.12 seconds with 70kV tube voltage and 7mA. Peri-apical radiographs were developed with an image plate scanner (Dürr Dental VistaScan Mini Plus, Bietichheim-Bissingen, Germany).

All two-dimensional radiographic images were viewed in full-screen modus with specialized imaging software (Mediadent, Image Level version 6.14.4.24F, Kruibeke, Belgium) on a 24-inch monitor (Barco Eonis® MDRC-2224 BL, Kortrijk, Belgium).

Statistical analysis

Data analysis was performed in IBM SPSS® Statistics 25 with the furcation site as the statistical unit. Descriptive statistics included frequency distributions for categorical variables (gender, FI Grade) and mean values and standard deviations for continuous variables (age). The FI Grade as registered on CBCT was considered the true FI Grade (= gold standard). Absolute agreement and Cohen’s weighted kappa was calculated to determine the accuracy of panoramic and peri-apical radiography in assessing the FI Grade. FI Grades were recategorized into ‘no to limited FI’ (FI Grade 0 and I) and ‘advanced FI’ (FI Grade II and III) in order to determine the diagnostic sensitivity (SENS), specificity (SPEC), positive predictive value (PPV) and negative predictive value (NPV) of panoramic and peri-apical radiography in identifying advanced FI. For all diagnostic parameters 95% confidence intervals were calculated. To explore the possible impact of clinical experience, diagnostic parameters and area under ROC-curves were compared between experienced periodontists and postgraduate trainees using the Mann Whitney U-test. Finally, composite Receiver Operating Characteristic (ROC)-curves (25) were constructed with Python™ Software 3.8.0 to assess the diagnostic value of both imaging techniques on the basis of pooled data of all examiners, experienced clinicians or trainees. Therefore, the state value (CBCT) was recategorized into ‘no to limited FI’ (FI Grade 0 and I) and ‘advanced FI’ (FI Grade II and III). Test values included all FI Grades. The level of significance was set at 0.05.

Patient and site selection

In total, 60 furcation sites of 29 multi-rooted teeth from 17 patients (6 males, 10 females; mean age 62) could be included in this controlled clinical study. The original sample included 75 furcation sites, yet FI could not be assessed for 15 sites due to restoration materials causing CBCT artifacts. Fourteen of the 29 teeth were maxillary molars. Of those, 5 were first molars, 8 second molars and 1 third molar. Fifteen of the 29 teeth were mandibular molars. Of those, 5 were first molars, 8 second molars and 2 third molars. Twelve of the included furcation sites were mesiopalatal, 24 were buccal, 12 were distopalatal and 12 were lingual sites.

Examiners

Periodontists had a mean age of 43 years (SD 10.1) and included 6 males and 4 females. They had on average 16 years (SD 9.6) of clinical experience in private practice. Trainees had a mean age of 29 years (SD 1.9) and included 9 males and 1 female.

Assessment of FI on the basis of Cone Beam Computed Tomography (CBCT), panoramic and peri-apical radiographs

CBCT as gold standard

Based on detailed CBCT analysis, the degree of horizontal alveolar bone loss at the level of the furcation was classified as Grade 0 for 19 sites, Grade I for 25 sites, Grade II for 9 sites, Grade III for 7 sites.

Panoramic radiography versus CBCT

Details on all diagnostic parameters of panoramic radiography for assessing FI are shown in table 1. On average 19/60 (SD 4.2) furcation sites were correctly classified on the basis of panoramic radiography. This corresponded to weighted kappa of 0.209 (95% CI: 0.060–0.376), indicative of slight agreement. On average, 10/19 (SD 4.0) furcation sites with FI Grade 0 were correctly identified. For FI Grade I, II and III, the proportion of correct assessments amounted to 6/25 (SD 2.8), 2/9 (SD 1.0) and 2/7 (SD 2.2), respectively (table 2). Weighted kappa amounted to 0.231 (95% CI: 0.052–0.413) for experienced clinicians and 0.186 (95% CI: 0.067–0.338) for trainees. Weighted kappa amounted to 0.301 (95% CI: 0.020 – 0.525) in the maxilla and 0.238 (95% CI: 0.030 – 0.532) in the mandible. There was no significant difference between experienced clinicians and trainees, nor between the maxilla and mandible (P > 0.257).

When recategorizing FI into ‘no to limited FI’ (FI Grade 0 and I) and ‘advanced FI’ (FI Grade II and III), SENS of panoramic radiography amounted to 0.558 (95% CI: 0.490– 0.622). SPEC was considerably higher pointing to 0.791 (95% CI: 0.742–0.840). The results on PPV and NPV were 0.511 (95% CI: 0.409–0.612) and 0.804 (95% CI: 0.742–0.865), respectively. SENS was significantly lower for experienced clinicians than for trainees (P = 0.049). There was no significant difference between experienced clinicians and trainees for the other diagnostic parameters (P ≥ 0.082).

Composite ROC analysis showed an area under the curve of 0.79 for all examiners (figure 1). For experienced clinicians the area under the curve was 0.79 and for trainees it amounted to 0.77 (figure 2). There was no significant difference between experienced clinicians and trainees (P = 0.289).

Table 1

Diagnostic parameters of panoramic radiography for assessing FI
	Experienced clinicians	Trainees	P-value	All clinicians
Accuracy	0.231 (95% CI: 0.052–0.413) AUC: 0.79	0.186 (95% CI: 0.067–0.338) AUC: 0.77	0.257	0.209 (95% CI: 0.060–0.376) AUC: 0.79
Sensitivity*	0.487 (95% CI: 0.421–0.553)	0.624 (95% CI: 0.515–0.733)	0.049	0.558 (95% CI: 0.490–0.622)
Specificity*	0.842 (95% CI: 0.799–0.884)	0.740 (95% CI: 0.656–0.824)	0.082	0.791 (95% CI: 0.742–0.840)
Positive Predictive Value*	0.534 (95% CI: 0.436–0.631)	0.487 (95% CI: 0.286–0.689)	0.473	0.511 (95% CI: 0.409–0.612)
Negative Predictive Value*	0.796 (95% CI: 0.727–0.866)	0.811 (95% CI: 0.694–0.928)	0.940	0.804 (95% CI: 0.742–0.865)
*’No to limited FI’ = FI Grade 0 and FI Grade I; ‘Advanced FI’ = FI Grade II and FI Grade III
AUC Area Under (ROC)-Curve

Table 2

Correctly identified furcation involvements sorted per FI Grade and type of radiography
Examiner	FI Grade 0 (n = 19)		FI Grade I (n = 25)		FI Grade II (n = 9)		FI Grade III (n = 7)
Examiner	Pano	PA	Pano	PA	Pano	PA	Pano	PA
E(1)	10	7	6	1	0	0	5		4
E(2)	9	6	3	2	1	0	7		5
E(3)	11	13	2	4	2	1	1		3
E(4)	10	12	5	2	2	0	5		7
E(5)	13	12	6	8	4	1	0		2
E(6)	13	9	8	14	3	1	0		3
E(7)	11	13	3	4	3	3	2		-
E(8)	8	9	7	9	2	0	2		3
E(9)	13	13	4	2	2	0	1		1
E(10)	9	8	9	8	2	2	5		4
T(1)	13	10	6	10	1	1	0		5
T(2)	4	6	8	6	0	0	2		3
T(3)	11	11	8	3	2	0	2		2
T(4)	14	11	4	8	0	1	-		2
T(5)	1	3	8	2	1	1	0		5
T(6)	18	17	1	2	1	0	-		3
T(7)	10	10	5	4	1	0	2		-
T(8)	6	5	13	9	2	2	5		3
T(9)	8	8	6	8	2	1	2		1
T(10)	4	6	5	3	1	0	0		2
Mean	10	9	6	5	2	1	2		3
Pano Panoramic radiography
PA Peri-apical radiography
E Experienced clinician
T Trainee

Peri-apical radiography versus CBCT

Details on all diagnostic parameters of peri-apical radiography for assessing FI are shown in table 3. On average 19/60 (SD 4.2) furcation sites were correctly classified on the basis of peri-apical radiography. This corresponded to weighted kappa of 0.221 (95% CI: 0.051–0.404), indicative of slight agreement. On average, 9/19 (SD 3.4) furcation sites with FI Grade 0 were correctly identified. For FI Grade I, II and III, the proportion of correct assessments amounted to 5/25 (SD 3.6), 1/9 (SD 0.9) and 3/7 (SD 1.6), respectively (table 2). Weighted kappa amounted to 0.208 (95% CI: 0.047–0.389) for experienced clinicians and 0.215 (95% CI: 0.055–0.418) for trainees. Weighted kappa amounted to 0.257 (95% CI: 0.015 – 0.341) in the maxilla and 0.151 (95% CI: 0.015 – 0.341) in the mandible. There was no significant difference between experienced clinicians and trainees, nor between the maxilla and mandible (P > 0.461).

When recategorizing FI into ‘no to limited FI’ (FI Grade 0 and I) and ‘advanced FI’ (FI Grade II and III), SENS of peri-apical radiography amounted to 0.441 (95% CI: 0.386–0.497). SPEC was considerably higher pointing to 0.790 (95% CI: 0.768–0.812). The results on PPV and NPV were 0.439 (95% CI: 0.413–0.846) and 0.780 (95% CI: 0.748–0.835), respectively. There was no significant difference between experienced clinicians and trainees for any of the diagnostic parameters (P ≥ 0.425).

Composite ROC analysis showed an area under the curve of 0.69 for all examiners (figure 1). For experienced clinicians the area under the curve was 0.69 and for trainees it amounted to 0.63 (figure 3). There was no significant difference between experienced clinicians and trainees (P = 0.759). Figure 4 shows a clinical example of a maxillary molar with FI Grade III.

Table 3

Diagnostic parameters of peri-apical radiography for assessing FI
	Experienced clinicians	Trainees	P-value	All clinicians
Accuracy	0.208 (95% CI: 0.047–0.389) AUC: 0.69	0.215 (95% CI: 0.055–0.418) AUC: 0.63	0.880	0.221 (95% CI: 0.051–0.404) AUC: 0.69
Sensitivity*	0.441 (95% CI: 0.373–0.508)	0.442 (95% CI: 0.339–0.545)	0.425	0.441 (95% CI: 0.386–0.497)
Specificity*	0.793 (95% CI: 0.758–0.825)	0.788 (95% CI: 0.751–0.825)	0.820	0.790 (95% CI: 0.768–0.812)
Positive Predictive Value*	0.454 (95% CI: 0.383–0.525)	0.424 (95% CI: 0.313–0.535)	0.677	0.439 (95% CI: 0.413–0.846)
Negative Predictive Value*	0.783 (95% CI: 0.683–0.884)	0.776 (95% CI: 0.667–0.885)	0.596	0.780 (95% CI: 0.748–0.835)
*’No to limited FI’ = FI Grade 0 and FI Grade I; ‘Advanced FI’ = FI Grade II and FI Grade III
AUC Area Under (ROC)-Curve

The main objective of this study was to compare panoramic and peri-apical radiography with CBCT for detecting FI. Since panoramic and peri-apical radiographs are still taken regularly in daily practice, data on their accuracy is of pivotal relevance(26). Proper assessment of FI with a high degree of accuracy is also important for adequate treatment management of periodontitis patients(11).

FI is considered a locus minoris resistentiae in periodontal disease progression with high prevalence in periodontitis patients. Svärdström and Wennstrom(27) studied the distribution of furcation lesions and found that these lesions were most prevalent in the maxilla and more specific at distal sites of upper first molars. Furcation lesions could be found in more than 50% of chronic periodontitis patients older than 30 years. Every second molar was involved in patients older than 40 years.

Measuring clinical attachment level by horizontal probing is a standard tool in clinical periodontology. However, access to furcations is not always easy and furcations may be difficult to predict in architecture, number of walls and extent. An ideal method to identify FI is by reflecting a mucoperiosteal flap. Evidently, this is impossible for ethical reasons when there is no pathology or clear clinical indication for surgery. Intra-surgical registrations have been often used as gold standard in the diagnosis of periodontal defects with an indication for surgery(23)(26)(28). CBCT was used as gold standard in the present study since these were available in the context of planning implant surgery. CBCT has been shown to be a good gold standard in multiple studies describing high levels of agreement between pre-CBCT FIs and intra-surgical findings(29)(30). In addition, CBCT enables to visualize root characteristics such as fusions and proximities.

In the present study, only slight agreement was found between panoramic radiography and CBCT and between peri-apical radiography and CBCT. On the other hand, all diagnostic parameters need to be evaluated in detail before clinical recommendations can be made. Diagnostic parameters such as SENS, SPEC, PPV and NPV can only be calculated for dichotomous variables. For this purpose, FI Grades were recategorized into ‘no to limited FI’ (FI Grade 0 and I) and ‘advanced FI’ (FI Grade II and III). This recategorization makes sense from a clinical point of view since FI Grade 0 and I require no or limited non-surgical therapy, whereas FI Grade II and III need advanced surgical intervention. Panoramic and peri-apical radiography showed low SENS (0.550 and 0.441, respectively), yet high SPEC (0.791 and 0.790, respectively) for identifying advanced FI. These findings imply relatively high false negative ratings and low false positive ratings. In other words, advanced FI is frequently overlooked, but when it is identified on het basis of panoramic radiography or peri-apical radiography it is most likely present. This can be explained by the fact that panoramic and peri-apical radiographs are two-dimensional images of a three-dimensional anatomy. A certain overlap is to be expected. Even advanced lesions might be masked due to superimposition of bone, roots and restoration materials. On the other hand, peri-apical radiographs score superiorly for image quality (brightness, contrast) and bone details (quality of bone, contour of lamina dura)(31)(32). Panoramic radiographs provide a good overview, yet image distortion is an important limitation (6)(33). Interestingly however, is that panoramic radiography was not inferior to peri-apical radiography for detecting FI in this study. This may be explained by the fact that the quality of digital panoramic radiographs has been drastically enhanced in recent years. ROC-curves showed similar results with an even slightly higher area under the curve for panoramic radiography when compared to peri-apical radiography (0.79 versus 0.69).

The results on PPV and NPV are more difficult to interpret than SENS and SPEC since both are affected by the prevalence of advanced FI within the study sample. In this study, 16/60 furcations were Grade II or III. Only when this proportion resembles the proportion of advanced FI in the population, the data on PPV and NPV are valid. Clearly, the present study was based on a convenience sample of patients seeking implant therapy, which may not properly represent the population. In addition, substantial regional differences may exist in the prevalence of advanced FI among periodontitis patients.

In a retrospective study of Darby et al.(11), the diagnostic accuracy of furcation probing for detecting FI was investigated using CBCT as gold standard. Only 22% of the furcation sites were clinically accurate in grading compared with CBCT and 58% were overestimated. These findings indicate high SENS and low SPEC of furcation probing. Accuracy of furcation probing is dependent on factors like inclination and angulation of the probe, variability in operator’s technique/ inherent probing error, amount of force used when probing, tooth position, presence of adjacent teeth, restricted visualization of the probe due to limited mouth opening and difficult access to the entrance of the furcation. Also, the clinician can rather score the furcation concavity than the furcation itself. Deep root concavities may be confused with FI. All these factors may explain the high overestimation of clinical FI measurements. On the basis of the results of Darby et al.(11) and the results of the present study, it is clear that neither furcation probing, nor panoramic/peri-apical radiography are excellent examination methods on their own for the detection of FI. However combined, more furcations may be accurately assessed given the fact that furcation probing demonstrates high SENS whereas panoramic/peri-apical radiography show high SPEC. This is in line with a study of Gusmao et al.(34) and Greatz et al.(35) indicating that both furcation probing and radiographical assessment should be used in situations of suspected FI.

Especially when it comes to surgical decision making, it seems attractive to take a three-dimensional radiograph(8). Indeed, when teeth require complex periodontal therapy and also restorative treatment is necessary, CBCT might be a good additional tool for an accurate assessment and prognosis of multi-rooted teeth. Inaccurate diagnosis can lead to irreversible treatment planning decisions (11). Still, the diagnostic benefits of CBCT must be carefully balanced against a higher radiation dose(35). The radiation dose of a low-dose small-field CBCT is slightly higher when compared to a digital panoramic radiograph, yet substantially higher when compared to a peri-apical radiograph(11). When translated to clinical practice, a low-dose small-field CBCT might be considered when the architecture of the defect has a clear impact on the treatment strategy(36). This may apply to mandibular furcations with FI Grade II since these qualify for regenerative periodontal surgery.

An interesting finding was that diagnostic parameters for panoramic and peri-apical radiography were not affected by the experience of the clinician. Hence, gaining clinical experience does not seem to improve the accuracy of a radiological diagnosis of furcation sites.

When interpreting the results of the present study, a number of limitations should be taken in to account. First, this is a retrospective study based on a convenience sample. Second, we used the most common classification on FI because of its simplicity. On the other hand, FI also has a vertical component(37), which was not considered here. By classifying the vertical component, it could clinically influence the treatment strategy. Essentially, treatment of FI may involve periodontal regeneration, root resection, amputation, tunneling techniques and tooth extraction(38)(39)(40)(41)(42). Prognostically, the vertical component increases the risk of tooth loss significantly(43) and needs to be regarded as a risk factor in personalized maintenance programs(44). Third, CBCT was used as gold standard because of ethical restrictions. Assessing FI by means of intra-surgical examination remains the most accurate method(11).

In conclusion, panoramic and peri-apical radiography are relevant in the diagnosis of FI given high specificity. However, when clinical examination indicates pathology, diagnostic accuracy may be improved when clinical measurements are combined with radiographic examination. In clinical applications where CBCT images are present for justified reasons, the available CBCT images are more accurate than 2D images to assess furcation involvement. These are preferably combined with furcation probing, which shows high sensitivity. Clinical experience does not seem to improve the accuracy of a radiological diagnosis of furcation sites.

SPEC: Specificity

SENS: Sensitivity

PPV: Positive predictive value

NVP: Negative predictive value

ROC: Receiver Operating Characteristic

SD: Standard deviation

FI: Furcation involvement

CBCT: Cone-beam computer tomography

FOV: Field of view

Ethics approval and consent to participate
The study was conducted in accordance with the Helsinki declaration of 1975 as revised in 2000. The protocol was approved by the ethical committee of the Ghent University Hospital (B670201523577).

Consent for publication
Informed consent was obtained from all individual participants included in the study.

Availability of data and materials

The study was supported by the authors and their institutions. Patient demographics and radiographs were retrospectively selected from Ghent University Hospital records of patients who had been treated at the Department of Periodontology and Oral Implantology.

Competing interests

Gijs Berghuis declares that he has no conflict of interest.
Cosyn has a collaboration agreement with Nobel Biocare (Kloten, Switserland) and Straumann (Basel, Switserland).
Hugo De Bruyn has a collaboration agreement with Dentsply Sirona (Mölndal, Sweden) and Southern Implants (Irene, South-Africa).
Véronique Christiaens has a collaboration agreement with Southern Implants (Irene, South-Africa).

Funding
This study was funded by the authors and their institutions.

Authors’ contributions

GB performed data analysis for detecting furcation involvement, did interpretation of the results and was contributor in writing the manuscript. JC interpreted and contributed to the second and final draft as well as the design of the tables and statistical analysis. HB was first initiator of the research project. GH contributed to performance of radiological imaging and the final draft. MD performed the examination of FI based on CBCT. VC performed together with MD the examination of FI based on CBCT, was second initiator of the research project, collected data and contributed to the study design, first, second and final draft. All authors contributed significantly, read and approved the final manuscript.

Acknowledgements

Authors’ Information

Not applicable

(1) Tonetti MS, Greenwell H, Kornman KS. Staging and grading of periodontitis: Framework and proposal of a new classification and case definition. J Clin Periodontol 2018 June 01;45 Suppl 20:S149-S161.

(2) Caton JG, Armitage G, Berglundh T, Chapple ILC, Jepsen S, Kornman KS, et al. A new classification scheme for periodontal and peri-implant diseases and conditions - Introduction and key changes from the 1999 classification. J Periodontol 2018 June 01;89 Suppl 1:S1-S8.

(3) Tugnait A, Clerehugh V, Hirschmann PN. The usefulness of radiographs in diagnosis and management of periodontal diseases: a review. J Dent 2000 May 01;28(4):219-226.

(4) Papapanou PN, Tonetti MS. Diagnosis and epidemiology of periodontal osseous lesions. Periodontol 2000 2000 February 01;22:8-21.

(5) UPDEGRAVE WJ. The paralleling extension-cone technique in intraoral dental radiography. Oral Surg Oral Med Oral Pathol 1951 October 01;4(10):1250-1261.

(6) Mol A. Imaging methods in periodontology. Periodontol 2000 2004;34:34-48.

(7) Tugnait A, Clerehugh V, Hirschmann PN. Use of the basic periodontal examination and radiographs in the assessment of periodontal diseases in general dental practice. J Dent 2004 January 01;32(1):17-25.

(8) Walter C, Kaner D, Berndt DC, Weiger R, Zitzmann NU. Three-dimensional imaging as a pre-operative tool in decision making for furcation surgery. J Clin Periodontol 2009 March 01;36(3):250-257.

(9) Bower RC. Furcation morphology relative to periodontal treatment. Furcation entrance architecture. J Periodontol 1979 January 01;50(1):23-27.

(10) Van der Velden U, Abbas F, Armand S, Loos BG, Timmerman MF, Van der Weijden, G A, et al. Java project on periodontal diseases. The natural development of periodontitis: risk factors, risk predictors and risk determinants. J Clin Periodontol 2006 August 01;33(8):540-548.

(11) Darby I, Sanelli M, Shan S, Silver J, Singh A, Soedjono M, et al. Comparison of clinical and cone beam computed tomography measurements to diagnose furcation involvement. Int J Dent Hyg 2015 November 01;13(4):241-245.

(12) Hamp SE, Nyman S, Lindhe J. Periodontal treatment of multirooted teeth. Results after 5 years. J Clin Periodontol 1975 August 01;2(3):126-135.

(13) McGuire MK, Nunn ME. Prognosis versus actual outcome. III. The effectiveness of clinical parameters in accurately predicting tooth survival. J Periodontol 1996 July 01;67(7):666-674.

(14) Laky M, Majdalani S, Kapferer I, Frantal S, Gahleitner A, Moritz A, et al. Periodontal probing of dental furcations compared with diagnosis by low-dose computed tomography: a case series. J Periodontol 2013 December 01;84(12):1740-1746.

(15) Nordland P, Garrett S, Kiger R, Vanooteghem R, Hutchens LH, Egelberg J. The effect of plaque control and root debridement in molar teeth. J Clin Periodontol 1987 April 01;14(4):231-236.

(16) Cobb CM. Non-surgical pocket therapy: mechanical. Ann Periodontol 1996 November 01;1(1):443-490.

(17) Matuliene G, Pjetursson BE, Salvi GE, Schmidlin K, Bragger U, Zwahlen M, et al. Influence of residual pockets on progression of periodontitis and tooth loss: results after 11 years of maintenance. J Clin Periodontol 2008 August 01;35(8):685-695.

(18) Scarfe WC, Farman AG. What is cone-beam CT and how does it work? Dent Clin North Am 2008 October 01;52(4):707-30, v.

(19) Scarfe WC, Farman AG, Sukovic P. Clinical applications of cone-beam computed tomography in dental practice. J Can Dent Assoc 2006 February 01;72(1):75-80.

(20) Sukovic P. Cone beam computed tomography in craniofacial imaging. Orthod Craniofac Res Invalid date Invalid date;6 Suppl 1:31-6; discussion 179.

(21) American Dental Association Council on Scientific Affairs. The use of cone-beam computed tomography in dentistry: an advisory statement from the American Dental Association Council on Scientific Affairs. J Am Dent Assoc 2012 August 01;143(8):899-902.

(22) Ludlow JB, Timothy R, Walker C, Hunter R, Benavides E, Samuelson DB, et al. Effective dose of dental CBCT-a meta analysis of published data and additional data for nine CBCT units. Dentomaxillofac Radiol 2015;44(1):20140197.

(23) Akesson L, Hakansson J, Rohlin M. Comparison of panoramic and intraoral radiography and pocket probing for the measurement of the marginal bone level. J Clin Periodontol 1992 May 01;19(5):326-332.

(24) Pepelassi EA, Tsiklakis K, Diamanti-Kipioti A. Radiographic detection and assessment of the periodontal endosseous defects. J Clin Periodontol 2000 April 01;27(4):224-230.

(25) Metz CE. Receiver operating characteristic analysis: a tool for the quantitative evaluation of observer performance and imaging systems. J Am Coll Radiol 2006 June 01;3(6):413-422.

(26) Christiaens V, De Bruyn H, De Vree H, Lamoral S, Jacobs R, Cosyn J. A controlled study on the accuracy and precision of intraoral radiography in assessing interproximal bone defect morphology around teeth and implants. Eur J Oral Implantol 2018;11(3):361-367.

(27) Svardstrom G, Wennstrom JL. Prevalence of furcation involvements in patients referred for periodontal treatment. J Clin Periodontol 1996 December 01;23(12):1093-1099.

(28) Christiaens V, De Bruyn H, Thevissen E, Koole S, Dierens M, Cosyn J. Assessment of periodontal bone level revisited: a controlled study on the diagnostic accuracy of clinical evaluation methods and intra-oral radiography. Clin Oral Investig 2018 January 01;22(1):425-431.

(29) Walter C, Weiger R, Zitzmann NU. Accuracy of three-dimensional imaging in assessing maxillary molar furcation involvement. J Clin Periodontol 2010 May 01;37(5):436-441.

(30) Qiao J, Wang S, Duan J, Zhang Y, Qiu Y, Sun C, et al. The accuracy of cone-beam computed tomography in assessing maxillary molar furcation involvement. J Clin Periodontol 2014 March 01;41(3):269-274.

(31) Vandenberghe B, Jacobs R, Yang J. Diagnostic validity (or acuity) of 2D CCD versus 3D CBCT-images for assessing periodontal breakdown. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2007 September 01;104(3):395-401.

(32) Mengel R, Candir M, Shiratori K, Flores-de-Jacoby L. Digital volume tomography in the diagnosis of periodontal defects: an in vitro study on native pig and human mandibles. J Periodontol 2005 May 01;76(5):665-673.

(33) Eley BM, Cox SW. Advances in periodontal diagnosis. 1. Traditional clinical methods of diagnosis. Br Dent J 1998 January 10;184(1):12-16.

(34) Gusmao ES, Picarte AC, Ben Barbosa MB, Rosing CK, Cimoes R. Correlation between clinical and radiographic findings on the occurrence of furcation involvement in patients with periodontitis. Indian J Dent Res 2014 October 01;25(5):572-575.

(35) Graetz C, Plaumann A, Wiebe JF, Springer C, Salzer S, Dorfer CE. Periodontal probing versus radiographs for the diagnosis of furcation involvement. J Periodontol 2014 October 01;85(10):1371-1379.

(36) Avila-Ortiz G, De Buitrago JG, Reddy MS. Periodontal regeneration - furcation defects: a systematic review from the AAP Regeneration Workshop. J Periodontol 2015 February 01;86(2 Suppl):S108-30.

(37) Tonetti MS, Christiansen AL, Cortellini P. Vertical subclassification predicts survival of molars with class II furcation involvement during supportive periodontal care. J Clin Periodontol 2017 November 01;44(11):1140-1144.

(38) Sanz M, Jepsen K, Eickholz P, Jepsen S. Clinical concepts for regenerative therapy in furcations. Periodontol 2000 2015 June 01;68(1):308-332.

(39) Cortellini P, Cortellini S, Tonetti MS. Papilla preservation flaps for periodontal regeneration of molars severely compromised by combined furcation and intrabony defects: Retrospective analysis of a registry-based cohort. J Periodontol 2020 February 01;91(2):165-173.

(40) Mokbel N, Kassir AR, Naaman N, Megarbane JM. Root Resection and Hemisection Revisited. Part I: A Systematic Review. Int J Periodontics Restorative Dent 2019 February 01;39(1):e11-e31.

(41) Carnevale G, Pontoriero R, Hurzeler MB. Management of furcation involvement. Periodontol 2000 1995 October 01;9:69-89.

(42) Rudiger SG, Dahlen G, Emilson CG. The furcation tunnel preparation-A prospective 5-year follow-up study. J Clin Periodontol 2019 June 01;46(6):659-668.

(43) Nibali L, Sun C, Akcali A, Yeh YC, Tu YK, Donos N. The effect of horizontal and vertical furcation involvement on molar survival: A retrospective study. J Clin Periodontol 2018 March 01;45(3):373-381.

(44) Lang NP, Suvan JE, Tonetti MS. Risk factor assessment tools for the prevention of periodontitis progression a systematic review. J Clin Periodontol 2015 April 01;42 Suppl 16:S59-70.

Download PDF

Journal Publication

published 12 Mar, 2021

Read the published version in BMC Oral Health →

Editorial decision: Major revision
20 Dec, 2020
Review #2 received at journal
04 Dec, 2020
Review #1 received at journal
15 Nov, 2020
Reviewer #2 agreed at journal
13 Nov, 2020
Reviewer #1 agreed at journal
10 Nov, 2020
Reviewers invited by journal
08 Nov, 2020
Editor assigned by journal
03 Nov, 2020
Submission checks completed at journal
03 Nov, 2020
Editor invited by journal
03 Nov, 2020

You are reading this latest preprint version

A controlled study on the diagnostic accuracy of panoramic and peri-apical radiography for detecting furcation involvement

Status:

Journal Publication

Version 2

Abstract

Figures

Background

Methods

Results

Discussion

Conclusions

Abbreviations

Declarations

References

Status:

Journal Publication

Version 2