INTRODUCTION

Understanding dental caries, its progression and proper diagnosis in early stages has become a challenge in clinical practice, research, and prevention.¹

Tooth decay is a dynamic disease affecting the tooth’s surface in contact with biofilm, resulting in mineral loss, which leads to the localized destruction of the tooth’s hard tissue. During the early stages of the carious lesions there is an increase in enamel micro-porosity, which clinically appears as opacity or a stain that can be either white or brown. As micro-porosity increases, tooth structure collapses forming a cavity. This process is slow, and according to the evidence it can be stopped before the cavity appears.¹ However, while currently there is a decrease in caries prevalence nationwide, the percentage of this disease is still high, as reported in the 2014 National Oral Health Survey in Colombia (Estudio Nacional de Salud Bucal en Colombia -ENSAB IV), which showed that caries prevalence at age 5 was 52.38%, at age 12 it was 37.4%, and at age 15 it was 44.5%.²

The carious process is highly variable, with periods of progression alternating with periods of stability; therefore, diagnostic methods must seek to detect the aforementioned changes, in order to determine the stage at which the disease can be arrested. In consequence, the development of an effective diagnostic method has been the goal of many researchers for a long time.^1,3,4

Conventional caries diagnostic approaches include the visual-tactile method and the radiographic method, which are relatively accurate^3,4,5 but have been modified to improve effectiveness. Therefore, within the visual-tactile method there have multiple systems of caries classification, which tend to report the initial carious lesions due to changes in their pattern of presentation.

The International Caries Classification and Management System (ICCMS^TM) uses a simplified version of the International Caries Detection and Assessment System (ICDAS). ⁶ In an attempt to improve traditional methods, unify concepts, establish treatment decisions, and better detect this type of lesions, mainly at the interproximal level, several authors recommend a combination of the visual examination with other diagnostic aids. The most commonly used complementary technique is coronal x-rays or bite-wing x-rays, which is very helpful in detecting proximal lesions.^6,7,8,9,10

X-rays are useful to detect advanced tooth decay, since the demineralized tooth area allows a greater filtration of x-rays, showing as a radiolucent zone. The opposite occurs in early lesions, which do not have sufficient demineralization to be detected through x-ray. On the other hand, intraoral radiography may reveal interproximal caries, which could go unnoticed by clinicians during regular examination, but early lesions are difficult to detect by x-ray methods, especially if they are located in enamel only. Even experienced clinicians may disagree on the presence of decay by x-ray methods. Some can identify lesions on intact surfaces (false positives), while others fail to detect evident lesions (false negatives). This is why coronal x-ray as and additional diagnostic method is a useful complement to diagnose and accurately treat dental caries. In vitro studies have shown that this is the most reliable method to diagnose caries in the dentin middle third, compared with just visual inspection.^4,5

Like visual diagnostic methods, the radiographic technique has undergone changes thanks to technological advances. The conventional technique with x-ray film used to be popular in the past, while today digitized media is available, yielding better results by increasing image contrast. The 1991 studies by Wenzel et al found out increased effectiveness in digital radiographs, concluding that this technique improves diagnosis, compared with visual inspection, as it increases sensitivity without losing specificity.^7,11

The sensitivity of radiographic images reported in several studies is low; however, it remains the complementary technique of choice for most clinicians. The 2011 study by Diniz et al reported 44% sensitivity and 94% specificity.¹²

Currently, coronal x-rays are recommended for patients over 5 years of age, in order to reduce the underreporting occurring in interproximal lesions. Radiographically, the ICCMS^TM classifies posterior tooth surfaces in various degrees according to progression, enabling good reproducibility and accuracy of this classification system.^13,14

The ICCMS^TM integrated the ICDAS evaluation with the proposed radiographic evaluation to detect caries lesions and their depth, in order to provide more comprehensive information to plan, manage, and properly diagnose tooth decay.^15,16,17,18

The goal of this study was to establish correlation between clinical and radiographic diagnosis of dental caries as proposed by the ICCMS^TM, in deciduous and permanent molars in a school population.

MATERIALS AND METHODS

A descriptive study was conducted using the clinical and radiographic database of 35 kids (20 girls and 15 boys) aged 6 to 8 years, who were treated in a public school of the city of Cali, Colombia. The sample was chosen for convenience and the analysis unit consisted of 1260 permanent and deciduous molar surfaces. The surfaces had been previously evaluated for caries and were entered in each patient’s medical record following the ICCMS^TM, which regroups the ICDAS criteria in four categories: 0 = healthy; A = initial decay in enamel, showing a white and/or brown lesion spot in either dry or wet surface (ICDAS codes 1 and 2); B = moderate decay, including the micro-cavitated lesion and the dentine lesion with an underlying shade (ICDAS codes 3 and 4), and C = severe caries, cavitated lesions with visible dentin (ICDAS codes 5 and 6). The data were entered by a dentist standardized in this index, with a 0.80 Kappa index.

Patients had bitewing x-rays taken at Universidad del Valle, all by a single assistant using the X digital GX 770 equipment (Intra Oral X-Ray System/ User Manual Dentsplay-Gendex). The x-rays were analyzed and entered in a form for caries reading and classification by the ICCMS^TM. The radiographic criteria included were: R: 0 Not Radiolucent, RA: early stage, which includes: RA1 (radiolucency in the ½ outer enamel), RA2 (radiolucency in the ½ internal enamel ± ADJ (amelodentinal junction), RA3 (radiolucency limited to the ⅓ outer dentine). RB: mild stage, which includes RB4 (radiolucency that reaches the ⅓ half of the dentin), RC: extensive stage, which includes RC5 (radiolucency reaching the ⅓ inner dentin, clinically cavitated), and RC6 (pulp radiolucency, clinically cavitated). The following data were taken into account: occlusal, mesial, and distal surfaces of first and second deciduous molars and first permanent molars, for a total of 36 surfaces per patient. The analysis was performed by two dentists previously standardized as inter- and intra- examiner, in the radiographic system proposed by ICCMS^TM, with a 0.80 Kappa index.¹³

The inclusion criteria were: mesial, occlusal, and distal surfaces of deciduous and permanent molars that were entered as part of the clinical and radiographic analysis of the medical records. The exclusion criteria were: molar surfaces that were not present in the x-ray or the odontogram, or whose crowns were not erupted in at least ⅔; 86 surfaces were excluded accordingly, leaving 1174 surfaces to analyze. Due to the greater applicability of bitewing x-rays to evaluate interproximal surfaces, a new analysis was performed with just the interproximal surfaces (806 surfaces).

RESULTS

The clinical and radiographic diagnosis was correlated in 1174 coronal and proximal surfaces with 0.41 Spearman’s correlation coefficient (p < 0.05). Table 1 lists the results of the clinical and radiographic correlation of all surfaces, showing clinical and radiographic coincidence of a healthy diagnosis in 95.9% of cases. In early stages of tooth decay, there was 8.16% coincidence; however, 89.12% of clinically detectable lesions in early stages were recorded as healthy from a radiographic perspective.

The clinical diagnosis of mild lesions showed that coincidence with the moderate radiographic level was 6.45%. Concerning lesions clinically diagnosed as severe, 40.58% showed coincidence with the radiographic diagnosis, while 31.88% of these lesions were recorded as healthy from a radiographic perspective.

Table 2 shows the results excluding the analysis of occlusal surfaces, leaving a total of 806 interproximal surfaces with 0.48 Spearman’s correlation coefficient (p < 0.05). There was also a higher proportion of surfaces clinically and radiographically diagnosed in early stages (17.50%); similarly, the percentage of coincidence in mild stages was 16.67%, and there was 82.35% coincidence in severe stages.

DISCUSSION

Caries detection by radiographic means is only possible if the amount of demineralization is such that produces a change in radiation density. Bitewing x-rays have proved to be very useful to detect caries in interproximal and occlusal surfaces of deciduous and permanent molars, since direct visual inspection is not possible in all cases, as interproximal surfaces are wide and their contacts prevent correct diagnosis.^19,20 This is why x-rays as a diagnostic complement is more important in clinical than in epidemiological terms.

The present study showed low correlation between the caries categories proposed by the ICCMS^TM in relation to the radiographic system proposed, in analyzing both proximal and occlusal surfaces and interproximal surfaces only; this is due perhaps to the minimal mineral loss in early lesions, which cannot be radiographically perceived. In addition, the overlying of enamel in the occlusal area can mask the mineral loss process.

The DMF index is the international standard that the WHO has been using for the control of dental caries. This index gives an average of decayed (cavitated), missed, and filled teeth, providing a report of affected teeth or surfaces. In order to achieve early detection of caries in early stages, multiple systems of caries classification have been created, and while aiming at early diagnosis, they require specific clinical conditions and the clinician’s standardization. The ICCMS^TM system proposes a series of combined caries categories, based on the ICDAS classification system, sorting them out according to depth-which is an effective alternative in both clinical practice and education, research, and epidemiology-.⁶ In addition, for a comprehensive diagnosis it proposes the ICDAS/ ICCMS^TM radiographic system, categorizing lesions according to radiographic depth.

The present study showed that when lesions were clinically classified as initial according to the ICCMS^TM, 82.61% were radiographically reported as healthy. Again, the difficulty in radiographic diagnosis of early lesions may be related to the percentage of demineralization, as it has been reported that 30 to 40% of mineral loss is necessary for the radiographic detection of enamel caries.^21,22,23

In lesions classified as mild per the ICCMS^TM, 61.29% were radiographically classified as healthy. These results agree with the study by Bertella et al in 2012 on the x-ray pattern of mild lesions on permanent molars, evaluating the associationbetweenenamel fissure and radiographic features, finding out that 67.4% of these lesions did not show radiographic radiolucity.^12,22

A second analysis was conducted excluding the occlusal surfaces, since coronal x-rays have proven to be more effective in interproximal surfaces. This analysis yielded a greater coincidence of surfaces that are clinically and radiographically diagnosed in early and mild stages. This change may be explained because the use of bitewing x-rays for occlusal caries diagnosis has been questioned, since they have been considered more useful for interproximal lesions. Nevertheless, the study by Ricketts et al found out that x-rays are better than visual examination for the diagnosis of occlusal caries only. However, the use of bitewing x-rays to diagnose these lesions should be considered as a complement to visual diagnosis instead of a highly sensitive diagnostic aid.^5,13,14

Research results indicate that bitewing x-rays could add substantial information to the visual examination, allowing clinicians make better treatment decisions. Similarly, Rodrigues et al, in 2008, concluded that the clinical classification system combined with x-rays improved diagnosis, and suggested this combination for caries diagnosis.^12,23

Other studies, such as the one by Lobo et al, have shown that bitewing x-rays can add information to the visual examination concerning the stage of the carious process in more advanced stages, including hidden cavities.^12,24 Similar results were obtained in the present study, where extensive cavities coincided in a higher percentage with the radiographic diagnosis, indicating that bitewing x-ray examination provides more accurate information in advanced stages. In 1991, Tveit et al found out that x-rays were more successful in the diagnosis of dentine caries, where 100% of deep lesions were correctly diagnosed.²⁵

In the same way, Mitropoulos et al, in 2010, noted that visual inspection was more accurate than x-rays in the detection of initial enamel lesions, while the opposite happened in extensive lesions, which may indicate that x-rays are more useful to diagnose deep lesions.¹⁸

The present study makes useful contributions to the knowledge on the use of radiographs as a diagnosis method accompanied by the clinical diagnosis proposed by the ICCMS^TM, recognizing the importance of x-rays to define the depth of dentin lesions, rather than using them for early diagnosis.

Within the limitations of this study is the fact that the results are applicable exclusively to the evaluated population, and that the sample size, while being for convenience, fails to represent the entire population of this age; in addition, it has the limitations of this type of design. Further research is required to evaluate diagnosis and to make comparison with different methods, as well as studies on clinical and radiographic classification, helping the decision making process. Additional studies on clinical and radiographic correlation are also recommended, using different clinical and radiographic classification systems, using coronal x-ray in order to determine the most useful and accurate method in the clinical practice.

CONCLUSIONS

Based on the results of this study, the following conclusions can be drawn: