A Critical Evaluation of the Role of Radiography in the Diagnosis and Treatment of Dental Caries

Introduction
Dental caries, along with periodontal disease are the most prevalent oral diseases. Dental caries simply means rotting or decay of the teeth, and if left unchecked results in pain, discomfort and the eventual loss of the tooth. Caries lesions are the result of an imbalance in physiological equilibrium between tooth mineral and the biofilm fluid surrounding it. If biofilm or plaque is allowed to build up on the tooth surface, the resulting fall in pH causes demineralisation of the tooth surface. If the situation persists, this will result in the softening of the enamel structure, and the eventual loss of tooth tissue. It is estimated that each year in England alone, one million nights sleep are lost, and five million days disturbed by toothache caused by caries [1:5]. The prevention and treatment of dental caries is, therefore, of great importance to the dental profession. Accurate and timely diagnosis helps the clinician to provide the best treatment for the patient. Early carious lesions can often be controlled with changes in diet and attention to oral hygiene, together with the application of topical fluoride. However, once a cavity has formed, more invasive restorative treatment will be necessary. Radiography is an essential adjunct to clinical examination in the diagnosis of caries, but is more useful in some circumstances than others. This essay will examine the radiographic options available to the dentist in the diagnosis of dental caries, with reference to the aetiology and development of the disease.

Causes of caries development
The term dental caries is used to describe the signs and symptoms of a localised chemical dissolution of the tooth surface, caused by metabolic events taking place in the biofilm or plaque surrounding the tooth. Dental plaque forms on tooth surfaces which are not cleaned, and is readily apparent after two or three days without brushing [2:95]. Plaque is not food debris, but consists mainly of micro-organisms. The predominant theory of caries formation maintains that dietary carbohydrates are fermented by micro-organisms in the plaque on the tooth surface. Millar suggested that the fermentation process caused by bacteria in the plaque produces organic acids which lower the pH at the enamel surface below the critical pH (5.5) necessary to maintain mineralisation of the tooth. The enamel begins to dissolve, however as the carbohydrates are used up, the acid formed dissipates allowing remineralisation to occur [2:4]. The saliva surrounding the teeth is supersaturated with minerals and thus the enamel apatite is readily remineralised [3:95]. Although this process leads to a constant state of flux, in a normal mouth the enamel surface is in a state of dynamic equilibrium with its surrounding environment. Dental caries occur when the pH levels remain low for too long or when conditions in the mouth prevent remineralisation. Diet plays a huge part in the production of acids in plaque. Two to three minutes after exposure to simple sugars, plaque pH falls from the normal average of 6.8 to around pH 5, well below the demineralisation threshold, and takes about 40 minutes to return to its normal value [2:95]. Saliva plays an important, but not fully understood, part in protecting the tooth enamel from demineralisation [1:61-62]. Saliva is readily released in response to food stimulus, especially those foods with a low pH. Fast flowing saliva is alkaline (pH7.5-8.0), and buffers the acids in the food and those formed in plaque. Saliva also contains high concentrations of calcium and phosphate which, in the presence of fluoride contribute to the remineralisation of the enamel surface after exposure to acid [2:97].

Process of caries development and radiographic appearance
Net mineral loss will, over a period of months or years, lead to visible lesions or ‘white spots’ on the tooth surface. White spot lesions indicate demineralisation just beneath the surface of the enamel, although the enamel surface is itself intact, thus protecting the tooth from bacterial attack. Although clinically visible, at this point the disease cannot be detected radiographically. In response to these early changes, reactionary dentine is laid down at the pulpal surface beneath the enamel lesion, and peritubular dentine seals off many of the tubules, thus protecting the pulp [1:14]. These changes may be visible radiographically as a sclerotic band [4:98-99]. At the same time, demineralisation of the dentine near the enamel-dentine junction occurs. At this stage the lesion can be arrested by a change in the oral environment and remineralisation can occur after application of topical fluoride and without mechanical intervention [1:14]. If the disease is allowed to progress, a cavity eventually forms in the enamel, allowing bacterial access to the dentine, and thus to the pulp, the pulp becomes inflamed in response to bacterial attack and causes pain. Eventually the pulp itself will become infected causing necrosis and possibly the spread of infection via the apex to beyond the tooth [1:14]. Radiographically, the enamel cavity appears as a small, well defined radiolucency. As the lesion spreads to the dentine it extends beneath the enamel and towards the pulp and the margin is ill defined. Then as it reaches the enamel dentine border the lesion takes on a triangular configuration with the base of the triangle spreading into the dentine along the border. The lesion then forms another triangular point as it progresses towards the pulp [5:501]. Eventually cavitation of the enamel and the dentine causes a change in the radiographic outline of the crown, although it is not possible to determine radiographically the point at which the pulp becomes exposed [4:98]. Caries is most likely to form on tooth surfaces which are not regularly cleaned; surfaces which are irregular or difficult to reach. Thus common sites for the formation of carious lesions are the occlusal surfaces of posterior teeth and proximal areas. Cervical and root caries are less common [4:98] as are caries on the smooth surfaces of the teeth. Proximal areas are, by their very nature, difficult to assess visually, and so other methods must be employed to aid in diagnosis and treatment. Occlusal caries are often ‘hidden‘ beneath an apparently intact enamel surface and present a different set of problems.

Conventional radiographic techniques available for the detection of caries
There are various radiographic techniques available to the dentist: Bitewing radiography requires a film or detector to be placed parallel to the posterior teeth using a holder or stick on tab, on which the patient bites. Ideally all the posterior teeth along with the alveolar ridge are visualised, and aproximal areas are revealed. Periapical radiography also ideally involves a holder. The film is placed parallel to the long axis of the tooth and both roots and crown are visualised. Caries can be imaged in this way, but only one or two teeth are seen per film. Dental Panoramic Tomography (DPT): These radiographs use a large cassette to hold the film, and are taken outside the mouth, using tomography to blur out structures buccal and lingual to a sharp focal plane. DPT has low sensitivity for caries and is not recommended for caries detection in children [2:45], however, Pendlebury [6:20] argues that a DPT may be appropriate for a patient who presents with a grossly neglected mouth, with multiple clinically determined carious lesions. Bimolar Lateral Oblique Mandible: These extraoral views are taken by laying the patient’s head against a radiographic cassette. The central radiation beam is directed from behind the angle of the patient’s jaw to image the teeth adjacent to the film. Although less sharp than a conventional intraoral bitewing radiograph, the technique is useful where small children, or patients with learning or behavioural difficulties are unable to tolerate intraoral films [2:45]. For more details on all these techniques refer to Essentials of Dental Radiography and Radiology by Eric Whaites [7]

Challenges in conventional radiography
Although these techniques are well researched and reliable, there are challenges to be faced when attempting accurate diagnosis of caries from radiographs. When attempting to diagnose caries using radiographs, the dentist must be aware that a radiograph is a two dimensional image of a three dimensional structure [8:516]. Geometric alignment of the x-ray beam may, in fact, mask a lesion, or can make an enamel lesion appear to have progressed into dentine. Incorrect horizontal angulation will cause the teeth to appear overlapped, potentially masking proximal lesions. Changes in exposure factors may also affect the appearance or size of a lesion. Carious lesions are usually larger clinically than they appear radiographically and very early lesions may not be visible at all [7:223-224, 9:407]. The appearance of lesions may be mimicked by cervical burn-out, an optical illusion caused by the difference in radiolucency of the enamel, the dentine and the alveolar bone at the neck of the teeth [7:220-221]. Viewing conditions have been shown to affect the accuracy of diagnosis of carious lesions [10:192-196], as has the experience of the reporting dentist [8:515-523].

Radiography as a diagnostic tool for proximal caries
Traditionally bitewing radiography has been seen as an essential adjunct to high quality visual examination of teeth in the detection of proximal caries in both primary and secondary dentition. A review undertaken in 1990 by Pitts and Kidd [11:44] of 29 research studies on the value of bitewing radiography, concluded that clinical examination alone rarely identifies more than 50% of proximal lesions when compared with the number diagnosed by both methods together. A subsequent literature review [12:5-16] highlighted the importance of bitewing radiography not only for monitoring the progress of active lesions, but in the ongoing monitoring of lesions which have been treated conservatively and may show radiographic evidence of remineralisation over time. The literature suggests, that the use of bitewing radiography increases the detection rate of proximal caries, particularly in lesions which are at or beyond the stage where operative treatment would be considered necessary [13:23-30] [14:336-340].

Radiography as a diagnostic tool for occlusal caries
Occlusal caries are easier to detect clinically than proximal caries, however opinion is divided on the efficacy of radiographic detection give further reading. The FGDP (UK) Selection Criteria for Dental Radiography [6:41] maintains that bitewing radiographs offer a significant yield in the detection of occlusal caries however, advances in non-radiographic detection techniques may change the situation in the future. Disease progression and examination frequency In accordance with NRPB recommendations that any medical radiation dose given must be kept as low as reasonably achievable (ALARA), all radiographs must produce a diagnostic yield and affect the subsequent management of the patient. Fejerskov [3:118] maintains that bitewing radiography only produces an increased diagnostic yield over clinical examination when cavitation is suspected and thus intervention is required. The Faculty of General Dental Practitioners (UK) (FGCDP) have produced a set of recommendations outlining good practice in the use of radiographs for the diagnosis of dental caries [6:41-51]. In summary, the taking of ‘routine’ radiographs without suspicion of caries development is not supported. Although bitewing radiographs have been shown to be valuable in charting the progress of disease, especially in the proximal areas, the intervals between examinations cannot be fixed, and should be judged according to the caries risk status of the patient at that particular time. It should be borne in mind that individuals can move in and out of risk categories over time. Children considered to be ‘high caries risk’ should have 6 monthly posterior bitewings taken, medium risk children should have annual bitewing radiographs and low risk children should be examined at 12-18 month intervals in the primary dentition phase and two yearly in the secondary dentition phase. Although this might suggest that children are examined where no visual-tactile evidence may exist, studies still show a significant diagnostic yield [6:45]. There is less evidence evaluating diagnostic yield of radiographs for caries in adults and the same recommendations apply as for children, whilst noting that lifestyle and behaviour changes in adult life may impact on the patient’s risk status. For instance, reduction in efficacy of brushing due to illness, loss of dexterity or psychological change may cause increased plaque build up. Reduction in saliva production due to the onset of Sjogren’s syndrome, or the side effects of medication may also place an individual into a higher risk category [6:46].

Radiography as a diagnostic tool for root caries
Root Caries is caused by bacterial plaque which forms on the cementum after gingival recession, and is thus seen mainly in older patients or those with advanced periodontal disease [1:176]. Root caries of the posterior teeth are visible on horizontal bitewing radiographs with a bone loss of less than 6mm or on vertical bitewing radiographs. Root caries on anterior teeth are best imaged using periapical radiographs and are usually described as ill defined and saucer-like. Cervical burnout may mimic the appearance of root caries, but true carious lesions can be distinguished by a absence of an image of the root edge, and the appearance of a diffuse rounded inner border where the tooth substance has been lost [15:283]. Radiography as a diagnostic tool for secondary caries Secondary or recurrent caries occurs immediately adjacent to a restoration. It may be caused by marginal leaking around the restoration, allowing plaque formation, or if the original lesion was not completely evacuated. Such lesions are a frequent cause of pulp necrosis. The radiographic appearance depends on the amount of demineralisation which has occurred and can often be obscured by the overlying restoration [15:285]. Bitewing radiography for posterior teeth and periapical radiography for anterior teeth can help in diagnosing the extent of the lesion, but Waites [7:222] points out that leakage of radiopaque ions from an amalgam into the surrounding tissues may cause the area beneath to appear radiolucent, mimicking secondary caries.

Radiography as a diagnostic tool for rampant caries
Nursing caries is a form of rampant caries occurring in very young children who have frequent or prolonged consumption of sugary fluids through a bottle or feeder cup [16:129-130]. In older patients rampant caries, usually seen on the smooth surfaces of the teeth, is caused by conditions which reduce the flow of saliva. This may occur following radiotherapy of the head and neck which affects the salivary glands, as a result of surgical treatment, or by the onset of Sjorgren’s syndrome. This has a causative relationship with caries development, but there are contradictory theories as to the reasons for this [1:61-62]. The radiographic appearance is characteristic, and shows radiolucent shadows appearing at the necks of the teeth most obviously on the mesial and distal aspects [15:286-288]. Panoramic radiography may be useful to identify which teeth require more detailed examination for example Periapical radiographs [6:20].

Advances in diagnostic techniques
Over the last decades there has been a general decline in the prevalence of caries, probably due to scientific advances in preventative dentistry. The greatest difficulty in caries detection is in the initial stage of enamel caries, when prompt remineralisation therapy may halt or even reverse the progress of the disease [17:431]. Various techniques are being developed to aid caries detection at this early stage, some involving ionising radiation.

Non-ionising tools
FOTI or fibreoptic transillumination has been used for many years as an adjunct to clinical examination and bitewing radiography [2:100, 6:49]. A very fine fibreoptic light source is placed in the interproximal area and resulting variations in the appearance of the enamel used to identify carious lesions. Laser light and infrared light emitting devices are also available and perform well in the assessment of early occlusal caries [18:640]. Devices which pass an electric current through the tooth to detect the impedance or conductance of the tooth material are also useful for detecting occlusal lesions which may be missed by conventional radiography.

Developments using ionising radiation
Digital imaging
Developments in computer technology have made it possible to produce a digital radiographic image. There are three ways a digital dental radiograph can be produced; by digitising a film produced by conventional methods using a flat bed scanner, by using a charge-coupled device connected to a computer (CCD) and by the use of photostimulable phosphor imaging plates (PSP). The primary advantage of digital radiography over conventional techniques is in image storage and manipulation. A digital image may be manipulated to optimise contrast in a particular area of interest, and then stored on a picture archiving system (PACS) allowing the image to be retrieved at will and shared with other professionals. Digitisation of conventional images allows storage on a PACS system, but requires a large number of processes to achieve the final image, and is dependent on great operator care to produce a high quality final image [19]. CCD and PSP systems have been shown in in vitro studies to have comparable results in terms of sensitivity in detecting caries [17:431-436, 20:900]. Although CCD and PSP systems require a smaller radiation dose to produce an acceptable image, Pendlebury [6:27] points out the apparent dose advantage gained with digital systems may well be outweighed; the sensors, especially for CCD are smaller, necessitating more exposures to cover a given number of teeth. The rigidity of the detectors make them more difficult to place in the mouth potentially leading to more repeat exposures as found by Farrier [20:900-907]. The exposure latitude for PSP’s is also a matter for concern, as an overexposure will still produce a high quality image, thus operators may not be aware they are using a much higher dose than necessary.

Computed Tomography
Tuned Aperture Computed Tomography (TACT) and Cone Beam Computed Tomography (CBCT) are developments of medical computed tomography (CT) which construct radiographic slices of anatomy using volume aquisition. As the acquisition method differs from that of conventional CT a much lower radiation dose is need to produce an image, making these modalities a feasible way of imaging teeth. They are routinely used for 3D imaging of tooth root structure and jaw pathology. The SedentexCT project [21:49-50] carried out a systematic literature review which suggests that CBCT offers similar diagnostic accuracy to bitewing radiography for proximal caries, and enhanced sensitivity but reduced specificity for occlusal caries. They report that the evidence does not, at this stage, support the use of CBCT for caries detection and diagnosis, but that further study into the detection of occlusal caries using this method could be beneficial. They recommend, however, that CBCT examinations performed for other purposes should be examined carefully for caries lesions.

Summary and Conclusion
The dentist’s first method of detection for caries is visual-tactile exploration. However, many tooth surfaces are difficult to visualise or probe, specifically proximal and occlusal surfaces. Bitewing radiography has for many years been the examination of choice to aid in proximal caries diagnosis. Many teeth can be imaged on one film, and the overall patient dose per examination is low. Using holders and a standardised technique, radiographs are readily reproducible and can be used to effectively monitor the progress of disease. Bitewing radiography does, however have its’ own limitations: Radiolucency is not apparent until a significant amount of demineralisation has occurred, and thus early lesions are not shown at all. Superimposition, and a two dimensional image mean that the exact site and extent of a lesion are difficult to determine. Variations in technique or exposure can affect the appearance and apparent size of a lesion [7:223]. Dentists often disagree on the presence of carious lesions on one particular film and one dentist looking at the same film on different days may detect a different number of lesions [15:272]. Although bitewing radiographs have good sensitivity and specificity for proximal caries, the results for occlusal caries are not so reliable and visual-tactile examination is more reliable. Periapical radiographs show similar specificity and sensitivity to bitewing radiographs, but fewer teeth are shown on each radiograph, necessitating more films to show all of the posterior teeth, and thus a higher radiation dose. Digital intraoral radiography has similar detection characteristics to conventional film radiography, and uses a lower radiation dose, however digital plates are not so comfortable for the patient, and may necessitate more repeat examinations. Panoramic radiographs (DPT) have low sensitivity and specificity for caries due to the geometry used in producing the radiograph, and are thus unsuitable for caries detection except, as mentioned, as an overview in patients with rampant caries. Fibreoptic transillumination (FOTI) and electric impedance tests may be more useful than intraoral radiography for detecting occlusal lesions. Recent developments in computed tomography (CBCT and TACT) have shown similar specificity and sensitivity to bitewing radiography, but at a much higher patient dose. Thus bitewing radiography, whether conventional film or digital, remains the only reliable low dose radiographic means of detecting proximal, and to a lesser extent occlusal, carious lesions which have progressed into dentine and which cannot be detected by clinical examination.

References
1. Silverstone LM, Johnson NW, Hardie JM, Williams RAD. Dental Caries. London: Macmillan; 1981.
2. Rugg-Gunn AJ. Dental Caries. In: Welbury RR, editor. Paediatric Dentistry. Oxford: Oxford University Press; 1997.
3. Fejerskov O, Kidd E. Dental Caries. The Disease and it's Clinical Management. Second Ed. ed. Munksgaard: Blackwell; 2008.
4. Browne RM, Edmondson HD, Rout PGJ. Atlas of Dental and Maxillofacial Radiology and Imaging. London: Mosby-Wolfe; 1995.
5. Senel B, Kamburoglu K, Ucok O, Yuksel SP, Ozen T, Avsever H. Diagnostic accuracy of different imaging modalities in detection of proximal caries. Dentomaxillofacial Radiology 2010;39(8):501-511.
6. Pendlebury ME, Horner K, Eaton KA, editors. Selection Criteria for Dental Radiography. London: Faculty of General Dental Practitioners (UK); 2004.
7. Whaites E. Essentials of Dental Radiography and Radiology. Edinburgh: Churchill Livingstone; 2002.
8. Diniz MB, Rodrigues JA, Neuhaus KW, Cordeiro RCL, Lussi A. Influence of examiner's clinical experience on the reproducibility and accuracy of radiographic examination in detecting occlusal caries. Clinical Oral Investigations 2010;14(5):515-523.
9. Langland OE, Langlais RP. Principles of Dental Imaging. Baltimore: Williams and Wilkins; 1997.
10. Hellen-Halme K, Lith A. Effect of ambient light level at the monitor surface on digital radiographic evaluation of approximal carious lesions, an in vitro study. Dentomaxillofacial Radiology 2012;41:192-196.
11. Kilpatrick NM. History, Examination and Treatment Planning. In: Welbury RR, editor. Paediatric Dentistry. Oxford: Oxford University Press; 1997.
12. Pitts NB. The use of bitewing radiographs in the management of dental caries: Scientific and practical considerations. Dentomaxillofacial Radiology 1996;25(1):5-16.
13. Newman B, Seow WK, Kazoullis S, Ford D, Holcombe T. Clinical detection of caries in the primary dentition with and without bitewing radiography. Australian Dental Journal 2009;54(1):23-30.
14. Poorterman JHG, Vermaire EH, Hoogstraten J. Value of bitewing radiographs for detecting approximal caries in 6-year-old children in the Netherlands. International Journal of Paediatric Dentistry 2010;20(5):336-340.
15. White SC, Pharoah MJ. Oral Radiology, Principles and Interpretation. St. Louis: Mosby; 2000.
16. Page J, Welbury RR. Operative Treatment of Dental Caries. In: Welbury RR, editor. Paediatric Dentistry. Oxford: Oxford University Press; 1997.
17. Pontual AA, de Melo DP, de Almeida SM, Boscolo FN, Haiter Neto F. Comparison of digital systems and conventional dental film for the detection of approximal enamel caries. Dentomaxillofacial Radiology 2010;39(7):431-436.
18. Neuhaus KW, Rodrigues JA, Hug I, Stich H, Lussi A. Performance of laser fluorescence devices, visual and radiographic examination for the detection of occlusal caries in primary molars. Clinical Oral Investigations 2011;15(5):635-641.
19. Xavier CRG, Araujo-Pires AC, Poleti ML, Rubira-Bullen IRF, Ferreira O, Capelozza ALA. Evaluation of proximal caries in images resulting from different modes of radiographic digitalization. Dentomaxillofacial Radiology 2011;40(6):338-343.
20. Farrier FL, Drage NA, Newcombe RG, Hayes SJ, Dummer PMH. A comparitive study of image quality and radiation exposure for dental radiographs produced using a charge-coupled device and a phosphor plate system. International Endodontic Journal 2009;42:900-907.
21. SedentexCT. Radiation Protection: Cone Beam CT For Dental and Maxillofacial Radiology. Evidence based Guidelines 2011. In: SedentexCT; 2011.