Performance evaluation of different observers in the interpretation of panoramic radiographs by the mandibular cortical index

OBJECTIVE: Mandibular Cortical Index (MCI) is useful in osteoporosis screening. This study aim was to verify the impact of different expertise levels in the observers’ performances of MCI evaluation in panoramic radiographs (PR). METHODS: Six participants were recruited: one Oral Radiology professor with previous experience (#1), one master’s degree student and radiologist with previous experience (#2), one PhD student and non-radiologist with previous experience (#3), one PhD student and radiologist with no prior experience (#4) and two master’s degree students with no prior experience (#5 and #6). Observers were trained to use the MCI before analyzing 150 PRs of patients with dual energy X-ray absorptiometry (DXA) examination. RESULTS: The Spearman’s Correlation test showed inverse and weak to moderate correlation between MCI and T-scores, and it varied was according to the observe that performed the evaluation. Observers #2 and #3 (with previous MCI experience) showed the highest agreement with the professor, followed by students #4, #5 and #6. CONCLUSION: Students’ expertise using the MCI increased the agreement with the professor regardless of whether the student was an oral radiology specialist; the MCI was correlated with the DXA in all the observer’s evaluations, which indicates that previous training for the MCI can enable any dentist to apply it in osteoporosis screening.


INTRODUCTION
Osteoporosis is an osteometabolic disease of chronic and systemic characteristics. Its progression results in the deterioration of bone microarchitecture and reduction of the bone mineral density (BMD) [1]. Osteoporosis affects mostly postmenopausal women [2], although it can also affect men [3] and patients with diseases in the bone metabolism [4]. The main consequences of osteoporosis are osteoporotic or low-impact fractures [5], whose more frequent sites are the hip and the forearm [6], Such fractures can progress to morbidity and mortality, and it represents high costs to the health system [7].
Currently, the most reliable tool for diagnosis of osteoporosis is the DXA examination which shows quantitative measurement of BMD [8]. However, considering the high cost of the exam, it is not always accessible in many countries [9,10]. Thus, many patients with risk of osteoporosis are not properly diagnosed or treated. A bone fracture can be considered an osteoporosis signal, but, when it occurs, it comes down to the diagnosis done belatedly [11]. Therefore, screening tools for osteoporosis were developed to identify potential individuals that need to be referred for DXA exam [12].
On the other hand, dental surgeons frequently request PRs for patients who initiate dental treatments [13,14] as a complementary exam to the initial diagnosis and/ or for case follow-ups [15]. PR is a low-cost and accessible exam [16]. Studies show that it may indicate osteoporotic changes in the mandible [17], notably morphological alterations in the mandibular cortex [18]. Because of that, several radiomorphometric indexes were developed [17,19] and among them, the MCI is highlighted [18].
The MCI is a qualitative index and was developed by Klemetti et al. in 1994 [18]. It evaluates morphological features (lacunar and linear erosions) of the endosteal margin and inferior cortex on the mandible that goes from the region of the mental foramen to the region of the lower second molars. Currently, the MCI is reported in the literature as a highly useful index in the screening of osteoporosis [1,[19][20][21][22][23].
The index proposed by Klemetti [18] consists of the following classification for the endosteal margin of the mandible: -C1: well delimited and demarcated endosteal margin with absence of radiolucent defects in lacunar or/and rectilinear morphology indicates erosions and signals patients with no need for the DXA exam; -C2: endosteal margin presenting lacunar defects or moderate radiolucent rectilinear defects evidences the possibility of reduced BMD (or osteopenia) and the need for further investigation using the DXA; -C3: endosteal margin and mandibular cortex severely eroded with multiple semilunar or linear defects strongly suggests a patient with osteoporosis and the need for a DXA exam. The MCI is easy to apply and does not require the use of any calculations, software, or specific programs for its execution; however, it is an index that requires good acuity and visual perception by the observer. Thus, previous orientation for the dental surgeon is necessary to apply it correctly [24][25][26][27]. It is known that professionals of different occupational fields and experience with the MCI shows distinct sensibility for its application. Therefore, this study aims to assess the MCI evaluation concordance level when used by professionals of different areas with and without previous experience with the index and its correlation with the DXA.

Observers and orientations
Six observers participated in this study: one university professor in oral radiology with previous experience using the MCI (#1); one radiologist, post-graduation student, with previous experience using the MCI (#2); one non-radiologist, post-graduation student with previous experience using the MCI (#3); one radiologist, post-graduation student with no prior experience using the MCI (#4) and two master's degree students with no prior experience using the MCI (#5 and # 6).
Initially, observers received basic information about the index and a didactic scheme based on the original article [20] as described in Figure 1. Patients signed a free and informed consent term, performed PR to start dental treatment in the Dentistry School and peripheral forearm DXA was done in the same day. Data collection was carried out with the approval of the ethics committee (FR358902 145/10). All PRs were performed using the same device (Kodak 8000, Eastman Kodak Company, Rochester, USA) and the peripheral DXAs were also performed using the same device (pDEXA, Norland, Norland Medical Systems, Inc., White Plains, New York, United States of America).
An inverse correlation was observed between the peripheral DXA values and the MCI values of each observer (p<0.001). In addition, a strong correlation was set considering the proximity of the R s value of observers with R s value of the more MCI experienced observer #1, while weaker correlation was set for those who had less previous experience with the MCI, as described in Table 2.

DISCUSSION
Radiomorphometric indexes, such as the MCI, were developed to use PRs in the detection of patients that Panoramic radiographs and the MCI | Munhoz et al. need to do DXA exams [17,18]. The MCI is an index of easy applicability; however, previous experience with its particularities positively affects the adequate application as a screening tool for patients that are susceptible to osteoporosis [24]. This study verified that professionals with previous experience using this index are more competent to detect patients with decreased BMD. However, professionals without previous experience using the MCI but with previous orientations about the index were also able to screen patients with low BMD.
Similar results were observed in previous studies in which general dental surgeons received orientation for the MCI and applied it to the screening of patients that needed to be referred for a DXA exam. As well, such professionals were able to identify patients with low BMD [5,24,29]. However, in a study with a reduced number of observers (three), an inability to correlate the DXA results with the MCI was verified [30].
Another previous finding was the higher correlation between the MCI and the DXA exam by previously oriented radiologists when compared to non-radiologist dentists [24,29]. In the present study, we observed a similar result: observer #4 (a radiologist with no previous experience using the MCI) had a better agreement with observer #1 (a university professor with previous experience using the MCI) as well as a stronger correlation between the MCI and the DXA exam than the observers #5 and #6 (nonradiologists with no previous experience using the MCI). This greater facility in MCI application may be because radiologists are trained to interpret radiographs [24,31] and are able to verify details in radiographs with greater acuity.
The agreement between the observers, when compared to observer #1 who was considered the most qualified in MCI assessment, showed significant variations. The lower agreement with #1 was the correlation between the DXA exam and the MCI. This may indicate that the practice of using the MCI improves the results. [25] Among the most experienced observers (#2 and #3), the radiologist observer (#2) presented the highest agreement with #1 and a more significant correlation when compared to #1. This fact demonstrates that the combination of previous experience and expertise in radiology can improve the application of the MCI.

CONCLUSION
Finally, according to the results of this study, we concluded that the MCI could and should be used by dentists, regardless their specialties. Professionals could be oriented to screen osteoporosis in PRs and refer their patients for DXA exams when necessary, achieving an early diagnosis in case of BMD reduction, and consequently, decreasing the risk of osteoporotic fractures, the main deleterious outcome of osteoporosis. Yet, what we have found is more experienced practitioners with the MCI present more facility in detecting patients with low BMD.