Cone beam computed tomography (CBCT) in periodontal diseases: a Systematic review based on the efficacy model

Background Periodontal diseases are prevalent among adult populations. Its diagnosis depends mainly on clinical findings supported by radiographic examinations. In previous decades, cone beam computed tomography has been introduced to the dental field. The aim of this study was to address the diagnostic efficacy of cone-beam computed tomographic (CBCT) imaging in periodontics based on a systematic search and analysis of the literature using the hierarchical efficacy model. Methods A systematic search of electronic databases such as PubMed, Scopus, Web of Science, and Cochrane was conducted in February 2019 to identify studies addressing the efficacy of CBCT imaging in Periodontics. The identified studies were subjected to pre-identified inclusion criteria followed by an analysis using a hierarchical model of efficacy (model) designed for an appraisal of the literature on diagnostic imaging modality. Four examiners performed the eligibility and quality assessment of relevant studies and consensus was reached in cases where disagreement occurred. Results The search resulted in 64 studies. Of these, 34 publications were allocated to the relevant level of efficacy and quality assessments wherever applicable. The overall diagnostic accuracy of the included studies showed a low or moderate risk of bias and applicability concerns in the use of CBCT. In addition, CBCT is accurate in identifying periodontal defects when compared to other modalities. The studies on the level of patient outcomes agreed that CBCT is a reliable tool for the assessment of outcomes after the treatment of periodontal defects. Conclusion CBCT was found to be beneficial and accurate in cases of infra-bony defects and furcation involvements.

. However, periodontitis could lead to bone resorption depending on its severity; hence, its diagnosis and treatment planning relies on clinical methods supported by radiographic imaging [15][16][17].
There are several risks to using clinical examination alone, which could prevent the accurate diagnosis of periodontitis, including gingival tissue consistency, inflammation severity, pressure while probing, probe size, probing angulation, and dental restoration existence [18,19]. In dental practice, practitioners routinely utilize conventional radiography such as periapical, bitewing, and panoramic x-ray to evaluate the bone loss and overall condition of the periodontal disease [18]. Nevertheless, the twodimensional x-ray has some limitations, mainly due to the overlapping of structures [20]. Thus, the detection of bone craters, inter-radicular bone loss, and lingual and buccal marginal bone loss necessitate the consideration of threedimensional radiography [17,[21][22][23][24].
Cone-beam computed tomography (CBCT) has been used frequently in the last two decades in dentomaxillofacial region [25]. It has many advantages compared to conventional computed tomography (CT) including low price, low radiation dose, and ease of accommodation at dental offices [25][26][27]. In addition, it has the ability to view the structures in three dimensions [28][29][30]. CBCT images of periodontal bone lesions offer a highly informative value. The spatial representation of the alveolar bone in all three planes has a significant role in periodontology, as treatment decisions and long-term prognosis rely on it [11]. Accordingly, it can play a potential role as an adjunct to clinical examination in the case of periodontal diseases [28,31,32].
Evidence-based dentistry aims to identify the best available evidence to justify the efficacy and use of any dental imaging or test in actual practice. Accordingly, Fryback and Thornbury came up with a hierarchal model of efficacy in the early nineties to sort out the best available evidence for a diagnostic tool [33].
There are several published studies on the role of CBCT in periodontal diseases in the literature [13][14][15].
However, the extent to which CBCT is efficient and accurate in the diagnosis, treatment planning, decisionmaking, and treatment outcomes of periodontal diseases remains ambiguous. On the path to routine use, especially under consideration of higher radiation exposure to patients, the gain in additional information of clinical relevance has to be explored and evaluated. Consequently, we conducted a systematic review to address the efficacy of CBCT in periodontal diseases.

Methods
This review was conducted based on guidelines from Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) [34] and guidance from the center for reviews and dissemination (CRD) for undertaking a systematic review in health care [35]. The eligibility criteria for inclusion and exclusion were set. Then, the included studies were assigned to the suitable level of efficacy. In the meantime, the review question was designed according to the PICO (Population, or Problem, Intervention or Exposure, Comparison, Outcome) element [36]. Finally, each study was evaluated for quality using the predetermined tool for quality assessment (QUADAS 2). The research question was defined as "what is the diagnostic efficacy of CBCT in individuals with periodontal diseases?"

Literature search:
Four databases PubMed, Scopus, Cochrane, and Web of Science were searched till February 2019 to identify the relevant studies. The search strategy is shown in Table 1.

Study retrieval:
The resultant studies were subjected to a duplicate check on the RefWorks database. The studies were then reviewed by four authors for relevance based on inclusion and exclusion criteria. After that, the studies meeting the eligibility criteria were assigned for full-text screening. Where uncertainty was present, discussions were conducted between the authors to reach an agreement on whether to include or exclude a study based on the predefined inclusion and exclusion criteria.
Data extraction & quality assessment: Finally, each of the selected studies was assigned for data extraction and analysis. After that, each study was allocated its suitable level of efficacy. A Revised Tool for the Quality Assessment of Diagnostic Accuracy Studies (QUADAS 2) was used for quality assessment. This tool contains four domains: patient selection, index test, reference standard, and flow and timing. Each domain is assessed in terms of risk of bias and the first three domains are assessed in terms of concerns regarding applicability. Signaling questions are included to help judge the risk of bias [37].

Studies allocation
The search strategy of the four databases yielded 1717 articles: PubMed 539, Scopus 746, Cochrane 71, and Web of Science 555. After a duplicate check using Ref-Works, the result came up to 1262. These were subjected to the title and abstract screening by the two authors. A set of 65 studies were linked to the full-text review. A total of 28 articles were excluded because they did not possess at least one of the inclusion criteria. Studies reported by [28,[38][39][40][41][42][43][44][45][46][47][48][49][50] were ex vivo studies and out of our review.
Plaque-induced periodontitis was not addressed, therefore, studies on that issue were excluded. In addition, studies that did not belong to any level of efficacy were disregarded [51][52][53][54][55]. Studies that addressed bone density conducted by Al Zahrani et al. [56] and bone coverage conducted by Ferriera et al. [57] were also excluded. Published studies by Evangelista et al. [58], Sun et al. [59], and Leung et al. [60] discussed only the naturally occurring dehiscence and fenestration, hence, they were disregarded. Studies reported by Goodarzy et al. [61] and Nagao et al. [62] were excluded because they did not include patients having periodontitis. The case report presented by Naitoh et al. [63] was disregarded as well. Studies published in languages other than English; reported by Deng et al. [64]) was excluded. Figure 1 shows the results for systematic reviews according to the PRISMA flow chart. Table 2 shows the studies that were included and their suitable efficacy level.   Quality assessment After allocating each study its suitable efficacy level, special tools of quality assessment were used for each one as described in the literature [37]. Technical efficacy studies: There was no study identified on this level of efficacy.
There were three studies that included a previously published systematic, manual search of the reference lists of the included articles [64,81,87], among which one study by Deng et al. [64] was found to be published in a Chinese language and hence excluded. Diagnostic thinking efficacy: Only one study was found to be on the level of diagnostic thinking efficacy [66]. The author investigated the effect of CBCT on the treatment decision-making after taking into consideration the clinical parameters. Therapeutic efficacy: In this level of efficacy, only one study, Pajnigara et al. [67], seemed relevant. Patient outcome efficacy: Our research resulted in eight studies in which CBCT was used to address the patients' outcomes in relation to periodontal disease. All of the studies are randomized clinical trials [68,72,75,77,79,82,85]. Table 4, the CASP (critical appraisal skills program) checklist, was used to assess outcomes.
Societal efficacy: Only one study was found to be relevant in this level of efficacy, Walter et al. [69]. The quality assessment was done using the QUADAS 2 tool.

Systematic reviews
The remaining six studies [6,70,73,78,80,83] were found to be systematic reviews for which the AMSTAR-   [97] was used. It is a popular instrument modified from the original AMSTAR, which contains 16 checklist questions. (Refer to Table 5). The two authors meticulously screened each study in order to give a suitable answer for each checklist question.

Discussion
Alveolar bone loss is considered a primary symptom of periodontal diseases. Mostly, the assessment and treatment decisions depend on clinical measurements supported by conventional imaging modalities. However, 2D imaging has its own limitations for detecting bone defects, including overlapping. An estimation of bone loss buccolingually has led to the consideration of 3D imaging. However, to what extent the CBCT is effective in the diagnosis of periodontal diseases is not yet clear. Accordingly, our systematic review was designed to summarize the available evidence according to the hierarchal model of efficacy developed by Fryback et al. [33]. In our systematic review, we decided to exclude studies that are published in any language other than English because of time restriction. In addition, case reports and narrative reviews are considered in the literature as lowevidence studies. Studies addressing periapical conditions and implant-related periodontal problems were also excluded as they are beyond our aspect in this review. In the meantime, it was decided to not include studies conducted ex vivo where the periodontal defects are created artificially since we believe those results will not mimic the CBCT's performance when conducted on humans.

Technical efficacy level
It seems most of the studies conducted on the use of CBCT in periodontal disease were aimed at performance detection, accuracy estimation, or the treatment outcome assessment. The authors found no study reported in the literature dealt with the technical aspect of CBCT.

Diagnostic accuracy level
As mentioned earlier in this review, the QUADAS 2 tool was used for the quality assessment of diagnostic accuracy studies. Only studies conducted in vivo were included in this review. Some studies did not use explicit reference standards to compare CBCT with other modalities [71,89,90,93,94].
Cimbaljevic et al. [63] compared the periodontal probing with CBCT in the terms of furcation involvement in the absence of a reference standard. Likewise, Darby et al. [64] addressed the discrepancies in the clinical measurements obtained from patients' records with their available CBCT images. A study conducted by Suphaanantachat et al. [92] compared CBCT to conventional intraoral radiography. However, they did not use an actual reference standard for comparison. Similarly, Zhu J. et al. [86] has focused on the reproducibility of the different parameters of CBCT for the furcation involvement evaluation, and hence, no reference standard was used.

Diagnostic thinking
A study published by Walter et al. [66] on decisionmaking revealed discrepancies between clinically and CBCT-based therapeutic treatment approaches. The discrepancy was found after 59-82% of the teeth were investigated to find out whether less invasive or most invasive treatment should be considered. However, they concluded that CBCT provides informative details in cases of furcation involvement, and hence, it is considered a reliable tool in decision-making regarding treatment of furcation involvement.

Therapeutic efficacy
According to our interpretation and in correlation with the hierarchical model of efficacy [33], we found that the study conducted by Pajnigara et al. [67] fits on this level. They investigated the pre and post-surgical measurements of clinical and CBCT for furcation defects. Although they reported statistically significant differences between; clinical-presurgery CBCT (P < 0.0001, 95% CI)    and clinical-post surgery CBCT; the three-dimensional imaging gives dental practitioners the chance to optimize treatment decisions and assess the degree of healing more effectively.

Patient's outcome efficacy
Our systematic review has revealed eight studies that used CBCT to assess the results of treatment provided for periodontal diseases [68,72,75,77,79,82,85,98]. However, it seems that this study is in disagreement with a previously published review [6]. They did not identify any study on the level of patient outcome. The reason for this could be the difference between our inclusion and exclusion criteria and theirs. All studies agreed that CBCT is a reliable tool in the assessment of the results of treatment using a bone graft.

Societal efficacy
The study reported by Walter et al. [69] has shown that the use of CBCT decreases the cost and time for periodontitis screening. However, CBCT should only be advised in cases of advanced therapy. Further studies with a sufficient number of patients were suggested.

Systematic reviews
Our review has resulted in six studies, which are systematic reviews. Each review is supposed to adhere to the criteria provided by AMSTAR and scores YES whenever applicable. The review published by Haas et al. [78] did not elaborate on whether they included the study registries or consulted content experts in the field in terms of comprehensive literature search strategy. Although a meta-analysis was conducted in such a review, the review authors did not assess the potential impact of risk of bias on the results of the meta-analysis or other evidence synthesis. Moreover, the authors did not carry out an adequate investigation of publication bias (smallstudy bias) or discuss its likely impact on the results of the review. Based on our interpretation, the study has not reported any source of funding or mentioned any conflict of interest. The study by Walter et al. [79] did not clearly have an explicit statement that the review methods were established prior to the conduct of the review and did not justify any significant deviations from the protocol. In addition, only one database has been searched for relevant studies. According to the AMSTAR2 criteria, the included studies were not described adequately. The study has not reported on the source of funding for the  individual studies included in the review. To our knowledge, the risk of bias has not been elaborated upon in the relevant sites in the review. Moreover, the review authors did not account for the risk of bias in individual studies when interpreting or discussing the results of the review. In addition, the authors have not reported any source of conflict including any funding they received for conducting the review. The review by Anter et al. addressed the accuracy of the CBCT as a tool for the measurement of alveolar bone loss in periodontal defects. However, the authors did not report that they followed PICO, which is a framework for review question formulation [36]. In terms of a comprehensive search strategy, we saw that this review did not fulfill the criteria regarding study registries and expert consultation in the field. Furthermore, the authors did not conduct the search in duplicate for the purpose of study selection. The review authors had also not performed data extraction in duplicates. According to our interpretation, the included studies were not described in appropriate detail. Additionally, the source of funding for each relevant individual study was not reported.
The study reported by Choi et al. [80] did not specify whether if there was a deviation from protocol, metaanalysis plan, or causes of heterogeneity if appropriate. In addition, a list of the excluded study in association with a justification for exclusion of each potential study has not been provided. Regardless of whether it is one of the targets of the review, this review has not discussed any potential risk of bias of the included studies. Moreover, the source of funding of each included study was also not reported. It could be included that this review does fulfill the AMSTAR2 [97] checklist to some extent.
The review by Woelber et al. [83] neither mentions any deviation from protocol whenever applicable nor elaborates on if is a plan for meta-analysis, if appropriate. In addition, a plan for investigating the possible causes, if appropriate, regarding heterogeneity was also not reported. The source of funding for each included study was not reported either. To some extent, the review fulfills the checklist of AMSTAR2.
According to our systematic review and AMSTAR2 tool, we found the review conducted by Nikolic-Jakoba et al. [6] best fulfills the tool criteria. However, the study's authors did not justify the reason for exclusion of each potentially relevant study from the review. As other reviews were included in our study, the source of funding of each included publication was not reported.

Conclusion
We concluded that most of the studies conducted on the rule of CBCT in periodontal diseases were at diagnostic accuracy level followed by the patient outcome level. Accordingly, it was found that CBCT is quite beneficial and accurate in the diagnosis of infra-bony defects and furcation involvement. Similarly, it is reliable in the assessment of the outcome of periodontal surgery and regenerative therapy. Furthermore, more studies with a larger cohort on the level of diagnostic thinking, therapeutic, and societal efficacy are needed to set up a clear guideline and evidence for the usefulness of CBCT.
Abbreviations CBCT: Cone beam computed tomography; CT: Computed tomography; PRISMA: Preferred items for systematic review and meta-analysis; CRD: Center for reviews and dissemination; PICO: Population, or Problem, Intervention or Exposure, Comparison (if appropriate), Outcome you would like to measure or achieve; QUADAS2: Quality assessment tool for diagnostic accuracy studies; CASP: Critical appraisal skills program; AMSTAR: A measurement tool to assess systematic review; 2D: Two-dimensional; 3D: Three-dimensional