NIH: Consensus Development Conference on Diagnosis and Management of Dental Caries Throughout Life: Background

Nigel B. Pitts, B.D.S., Ph.D., R.C.S., MFPHM:

I applaud the organizers for setting out an important and timely agenda for this conference, which is, in a sense, overdue. The focus of the conference is rightly on clinical practice and using current knowledge to provide the best possible care for individual patients. It is important to realize that much of the research in caries diagnosis has overlapped the applications of the diagnostic process in clinical practice, clinical research, and clinical dental epidemiology. The differing objectives, environments, and priorities of research in these areas often confuse attempts to synthesize the relevant literature, particularly when comparisons are being made across countries and cultures.

Since the aim of the conference is to develop scientifically based recommendations that can be applied by dentists and dental hygienists, it is important that the everyday fundamentals of clinical caries diagnosis are addressed clearly and objectively. Clinical diagnosis is the foundation on which the answers to most of the consensus questions will be based, either by providing information on caries detection or being used in the assessment of both primary and secondary preventive strategies as well as playing a key role in informing clinical decision-making. It is vital to consider the findings of the Research Triangle Institute (RTI) systematic review as well as those from other reviews from a variety of countries, even if some of the findings seem to contradict the dental facts of life taught to many of us and do not fit the "classical" findings of research carried out years ago. We would expect that clinicians in various countries may find different recommendations either easy or difficult to apply, and we should learn from the work done in medicine (SIGN, 1999) that there is also a developing literature on how to disseminate the findings of reviews effectively.

A key area is clarity about definitions and nomenclature. Many apparently similar terms are often used interchangeably in the literature but are taken by different researchers and clinicians to mean very different things. There will have to be clarity with regard to defining the terms "diagnosis" (not just detection), clinical "management" (encompassing preventive care of reversible lesions as well as surgical excision of tooth substance), "dental caries" (the view held for many years in Europe and now increasingly in the United States is that caries is a continuum rather than the macroscopic cavitation that is the late stage of the disease process), "throughout life" (here we need to differentiate early childhood caries from lesions in children, adolescents, adults, and seniors), and to plan minimally invasive care for the long-term benefit of the patient.

A European Perspective on the RTI Review

To make best use of the RTI review, it is important to understand the concepts of the D1 and D3 diagnostic thresholds used in it. Figure 1 shows an updated version of the iceberg analogy (Pitts, 1997a) for conceptualizing dental caries and the impact that a changing diagnostic threshold has on what is considered by dentists and researchers to constitute sound or diseased tooth tissue. The term "caries free" is frequently used when referring to data reported at the D3 (caries into dentin only) diagnostic threshold. This conveys the mistaken impression that there is no disease present, even though large numbers of carious lesions recognized as dental caries in the enamel are present (Pitts, Fyffe, 1988). The diagnosis of so-called "white spot" and "brown spot" caries has been accepted for many years in Europe and monitoring the behavior of these lesions over time is routine (Backer-Dirks, Amerongen, Winkler, 1951). It has been shown that the progression of these enamel lesions with macroscopically intact surfaces is extremely slow, and such lesions on free smooth surfaces do not always progress. They may stop, or even reverse (Backer-Dirks, 1966; Nielson, Pitts, 1991). These enamel lesions are often referred to as D1 lesions, as opposed to the D1 diagnostic threshold which includes both D1 and D3 lesions (see figure 1).

An example of the type of visual diagnostic criteria often used in European studies, which can be reported at either or both the D1 and D3 diagnostic thresholds, is the recently reported Dundee Selectable Threshold Method�(DSTM) (Fyffe, Deery, Nugent, et al., 2000a; Fyffe, Deery, Nugent, et al., 2000b). Traditional diagnostic aids (such as bitewing radiography and fiber-optic transillumination [FOTI]) detect more lesions still. The newer and more sensitive methods of caries diagnosis are now able to detect even more subclinical initial lesions which are in a state of dynamic progression and regression at an early stage of the disease process before they are discernible by conventional clinical methods. This gives the potential for lesions to be detected and the impact of preventive care to be assessed to ensure that cavitation is avoided.

The same iceberg can be used to link the diagnostic divisions of the continuum of dental caries with the type of management option that offers the patient the best long-term benefit. Choice of the most appropriate care option involves balancing the risk of continuing tooth destruction if preventive care fails against restorations placed and then replaced repeatedly over time with the imperfect methods currently available. The approach used in Europe for some years is summarized by the acronyms NAC for "No Active Care" above normal prevention, PCA for "Preventive Care Advised" when stable or noncavitated lesions are diagnosed, and PCA + OCA when both "Preventive and Operative Care Are Advised" for progressive dentinal lesions and lesions with significant cavitation (Pitts, Longbottom, 1995). There is a continuing debate in Europe as to exactly when restorative intervention is indicated, with movement toward recognizing the need to tailor the decision to the needs of individual patients and with a focus on cavitation rather than dentin involvement per se. It should be noted that hidden dentin lesions can sometimes be found in sites that are clinically sound, and that these lesions must be scheduled for operative care (see figure 2). It also must be emphasized that clinical caries diagnosed at the enamel lesion threshold with intact surfaces are not scheduled for restoration but are typically managed preventively in Europe.

Figure 1. Conceptualizing the caries process.

Figure 2. Linking diagnosis to clinical management.

A number of technical aspects of the RTI review are worthy of comment. The key finding that the quality of studies was often found to be poor may be seen as contentious by some in dentistry, and it is frustrating that (when measured against contemporary methodological standards) there are so few usable studies. However, it is important for these findings to be judged in the context of similar reviews in many fields of medical care where similar findings are common. They represent a major challenge to the dental research community.

Some areas of the review might have been improved if more time and resources had been available. A key concern in reviewing diagnostic literature in evidence-based healthcare is that the quality standards imposed in grading the papers are pertinent to the objective(s) of the study. Since data from some papers were employed for a number of different analyses (not always those intended by the authors), it might be argued that some of the quality scores were therefore inappropriate for some evidence tables. The presentation of the data is also complex. Other areas of debate include the possible use of receiver operating characteristic (ROC) analyses, rather than relying solely on sensitivity and specificity. Some argue that this method captures more of the diagnostic information obtained (ten Bosch, Mansson, 2000), while others are less convinced. Differences in the approach to histological validation are a further challenge. On the one hand, in vitro studies are commended as providing a true gold standard; on the other hand, differences between the diagnostic performance achieved in vitro and in vivo casts some doubt on the generalizibility of in vitro findings. The ideal study design (although very demanding in terms of logistics) would be to assess diagnosis in vivo first and then reassess the same surfaces in vitro following extraction of the tooth (for some ethically acceptable reason). A further difficulty occurs when the gold standard classically employed is potentially less sensitive than some of the methods being tested against it.

Studies Not Mentioned in the RTI Review

The papers cited below provide a European perspective on many of the challenges to clinical caries diagnosis raised in the review. The diagnostic challenge should not be underestimated or regarded as a basic or undemanding skill. The presentation of the disease has changed at a time when prevalence and incidence have slowed in some cases but become more polarized between risk groups (Kidd, Ricketts, Pitts, 1993) and as the range of preventive and operative treatment options has expanded (Paterson, Watts, Saunders, et al., 1991). Although clinical examination is the bedrock of daily dental practice, it is clear from many studies that clinical examination used alone will miss many lesions until they become so advanced that preventive intervention to avoid cavitation is compromised. The occlusal surface presents particular difficulties, since gross cavitation seems to occur less frequently and the limitations of the visual method have led to a fear of underdetecting hidden (or occult) lesions involving dentin.

A contentious issue for many clinicians is the lack of evidence supporting the continued use of a sharp explorer as a diagnostic tool. Although its use as part of a visuo-tactile clinical method is widespread and has been widely taught for many years in many countries, many European centers now teach that it is unethical to use an explorer in this way. This is because it was shown many years ago in Sweden (Bergman, Linden, 1969) that iatrogenic damage can readily be produced, particularly on initial caries within occlusal fissures, and favor continued lesion development. Similar findings were shown by Ekstrand and coworkers nearly 20 years later (Ekstrand, Qvist, Thylstrup, et al., 1987), when it was also shown experimentally that probing with an explorer had a deleterious effect in terms of subsequent enamel demineralization (Van Dorp, Ekterkate, ten Cate, 1988). The potential caries-causing damage was illustrated again by Yassin (1995). Apart from any risk of conveying cariogenic organisms from one fissure system to another, it is argued that a practice likely to cause harm to the patient cannot be justified if it fails to provide a significant balancing benefit. In this case, the absence of any diagnostic benefit from the visual + tactile method over the visual-only method means that the use of the sharp explorer for coronal caries diagnosis should be discontinued. A further complication in interpreting this literature is the difficulty of comparing studies which include open cavities in the assessment of occlusal caries diagnosis (Lussi, 1996).

My paper and presentation will include further elaboration of the content of relevant papers not found or not highlighted in the RTI review and the presentation of some new data. These references are listed below, following the draft recommendations.

1. What are the best methods for detecting early and advanced dental caries (validity and feasibility of traditional methods; validity and feasibility of emerging methods)?
   • Recognize that clinical caries diagnosis (with all its flaws) is the current foundation of lesion detection in clinical practice, clinical research, and clinical epidemiology. Care is needed to distinguish objective methods in each area.

   • Clinical visual methods of caries diagnosis are universally employed and are rapid, economical, and acceptable for detecting early-stage disease (enamel lesions, such as white and brown spot caries on accessible sites), noncavitated dentinal lesions, and late-stage cavitated caries. However, their inherent limitations must be remembered.

   • Although clinical diagnostic methods are highly specific, the low sensitivity achieved (particularly for noncavitated occlusal surfaces in vivo) means that the use of diagnostic aids with superior performance is indicated, and that new methods for caries diagnosis are needed.

   • Although the amount of high quality evidence on new diagnostic methods is less than desirable, the very limited evidence available on the efficacy of traditional diagnostic methods means that clinicians cannot be complacent.

   • Given the potential for caries-inducing and caries-accelerating iatrogenic damage from the use of a sharp explorer, combined with lack of any evidence of additional diagnostic benefit, sharp explorers should no longer be used for coronal caries diagnosis.

   • Educational initiatives will be needed to share the evidence on sharp explorers and persuade those still using them to give them up.

   • The long-term benefits to the patient of preventive caries management should be appreciated more readily by practicing dentists and should be the subject of continuing educational initiatives.

   • Scientific knowledge regarding caries diagnosis (and related preventive management) has moved ahead of many traditional professional, regulatory, and advisory frameworks.

   • The concepts of diagnostic thresholds should be more widely understood, and use of the ambiguous term "caries free" should be avoided.

   • It should be recognized that caries diagnosis in clinical practice, clinical research, and clinical dental epidemiology will have to change in light of continuing developments in knowledge. Strategies for systematically sifting, grading, and promoting new diagnostic approaches should be put in place internationally.

   • Attempts should be made to harmonize epidemiologic diagnostic methods in order to promote improved comparability and produce more reliable estimates of preventive care and restorative treatment needs.

2. How should clinical decisions regarding prevention and/or treatment be affected by detection methods and risk assessment?
   • There is a need for more reliable diagnostic methods to provide unambiguous indications of the extent, surface status, and activity of lesions.

   • There is a need for diagnostic methods that can reliably assess sealed surfaces.

   • There is a need for better tools for the diagnosis and treatment planning of secondary caries.

   • Before a decision to restore is made, clear evidence of significant cavitation or progressive dentinal involvement is needed.

   • Clinical diagnosis should lead into preventive-biased decision frameworks compatible with a PCA, PCA + OCA style of classification to avoid premature restoration of small noncavitated lesions.

   • There is a need for valid and reliable automated decision-support systems.

3. What are the promising new research directions for the prevention, diagnosis, and treatment of dental caries?
   • There is a need for more effective primary preventive products.

   • There is a need for secondary preventive products that can deliver lesion reversal prior to the cavitation stage.

   • There is an urgent need for high quality studies which are well conducted and well reported, using a minimum set of data meeting international standards.

   • There is a need for more studies evaluating the same lesions, both in vivo and in vitro.

   • There is a need for more studies of caries diagnosis in primary teeth.

   • There is a need for more studies evaluating diagnostic performance at the caries into enamel D1threshold.

   • There is a need for more studies on combinations of diagnostic methods with adjunctive and supplemental analyses.

   • There is a need for more sensitive, specific, and reliable diagnostic tools for early stage caries.

   • There is a need for diagnostic tools for lesions at the size where restorative intervention is indicated.

   • There is a need for diagnostic tools tailored for use in epidemiologic settings.

   • There is a need for diagnostic tools to detect hidden dentin caries.

   • There is a need for better restorative materials with physical properties more closely matching tooth tissue and able to act as a caries preventive agent when presented with a caries challenge.

   • There is also a need to develop the evidence base on how to disseminate effectively the findings of systematic reviews in dentistry and, having achieved that, how any changes in clinical practice which might be indicated can best be brought about.

References [(A) in parentheses denotes articles not found in the RTI review.]

(A) Axelsson P. Diagnosis and registration of carious lesions. In: Diagnosis and risk prediction of dental caries, Vol 2. Chicago: Quintessence, 2000.

(A) Backer-Dirks O. Longitudinal dental caries study in children 9-15 years of age. Arch Oral Biol 1961;6:94�108.

(A) Bergman G, Linden L. The action of the explorer on incipient caries. Svensk Tandlakare Tidskrift 1969;62:629�34.

(A) Deery CH, Care R, Chesters R, Huntington E, Stelmachonoka S, Gudkina Y. Prevalence of dental caries in Latvian 1- to 15-year -old children and the enhanced diagnostic yield of temporary tooth separation, foti and electronic caries measurement. Caries Res 2000;34:2�7.

(A) Deery CH, Fyffe HE, Nugent, ZJ, Nuttall NM, Pitts NB. General dental practitioners diagnostic and treatment decisions related to fissure sealed surfaces. J Dent 2000; 28:307�12.

(A) Deery CH. An evaluation of the use of pit and fissure sealants in the General Dental Service in Scotland. Ph.D. thesis, University of Dundee, Dundee, Scotland, 1997.

(A) Evans DJP, Matthews S, Pitts NB, Longbottom C, Nugent ZJ. A clinical evaluation of an Erbium:YAG laser for dental cavity preparation. Br Dent J 2000; 188: 677�9.

(A) Forgie A. Eyesight and magnification in dentistry. Ph.D. thesis, University of Dundee, Dundee, Scotland, 1999.

(A) Forgie AH, Paterson M, Pine CM, Pitts NB, Nugent ZJ. A randomised controlled trial of the caries preventive efficacy of a chlorhexidine containing varnish in high caries risk adolescents. Caries Res 2000;34:432�9.

(A) Fyffe HE, Deery CH, Nugent, ZJ, Nuttall NM, Pitts NB. Effect of diagnostic threshold on the validity and reliability of epidemiological caries diagnosis using the Dundee Selectable Threshold Method for caries diagnosis (DSTM). Comm Dent Oral Epidemiol 2000;28:42�51.

(A) Fyffe HE, Deery CH, Nugent, ZJ, Nuttall NM, Pitts NB. In vitro validity of the Dundee Selectable Threshold Method for caries diagnosis (DSTM). Comm Dent Oral Epidemiol 2000; 28:52�8.

(A) Kelly M, Steele J, Nuttall NM, Bradnock G, Morris J, Nunn J, et al. Eds: Walker A, Cooper I. Adult Dental Health Survey � Oral Health in the United Kingdom 1998. The Stationary Office, London, 2000.

(A) Kidd EA, Ricketts DNJ, Pitts NB. Occlusal caries diagnosis: A changing challenge for clinicians and epidemiologists. J Dent 1993;21:323�31.

(A) Longbottom C. The clinical diagnosis of dental caries � an initial examination of novel techniques. Ph.D. thesis, University of Dundee, Dundee, Scotland, 1992.

(A) Nugent ZJ, Pitts NB. Patterns of change and results overview 1985/6�1995/6 from the British Association for the Study of Comm Dentistry (BASCD) co-ordinated National Health Service surveys of caries prevalence. Comm Dental Health 1997;14(1)30-54. (Not In RTI bibliography, but results for 12-year olds for 96/97 were.)

(A) Paterson RC, Watts A, Saunders WP, Pitts NB. Modern concepts in the diagnosis and treatment of fissure caries. A review of clinical techniques and materials for the busy practitioner. London: Quintessence, 1991.

(A) Pendlebury M, Pitts NB, eds. Selection criteria in dental radiography. Faculty of General Dental Practitioners (UK), London, 1998.

(A) Pitts NB. Need for early caries detection methods: A European perspective. In: Stookey G, ed, Second International Conference on Detection of Early Caries. Bloomington, IN: Indiana University Press, 2000.

(A) Pitts NB, Deery C, Fyffe HE, Nugent ZJ. Caries prevalence surveys � a multi-country comparison of caries diagnostic criteria. Comm Dental Health 2000;17:196.

(A) Pitts NB. Diagnostic tools and measurements - impact on appropriate care. Comm Dent Oral Epidemiol 1997;25:24�35.

(A) Pitts NB. Patient caries status in the context of practical, Evidence-based management of the initial caries lesion. J Dental Education 1997;61:861�865.

(A) Pitts NB. The use of bitewing radiographs in the management of dental caries: scientific and practical considerations. Dentomaxillofac Radiol 1996;25:5�16.

(A) Pitts NB, Longbottom C. Preventive Care Advised (PCA) / Operative Care Advised (OCA)�categorizing caries by the management option. Comm Dent Oral Epidemiol 1995;23:55�9.

(A) Pitts NB, Longbottom C. Temporary tooth separation with special reference to the diagnosis and preventive management of equivocal approximal carious lesions. Quintessence Int 1987;18:563�73.

(A) Seddon RP. The detection of cavitation in carious approximal surfaces in vivo by tooth separation impression and scanning electron microscopy. J Dent 1989;17:117�20.

(A) SIGN Guideline: Targeted Caries Prevention in 6-16 year olds Attending for Dental Care. Scottish Inter-Collegiate Guideline Network, Edinburgh, December 2000.

(A) Sweeney PC, Nugent ZJ, Pitts NB. Deprivation and dental caries status of 5-year-old children in Scotland. Comm Dent Oral Epidemiol 1999;27:152�9.

(A) ten Bosch JJ, Mansson B. Characterization and validation of diagnostic methods. In: assessment of oral health, diagnostic techniques and validation criteria, ed. Faller RV. Karger, 2000. pp. 174�89.

(A) Van Dorp CSE, Exterkate RA, ten Cate JM. The effect of dental probing on subsequent enamel demineralization. J Dent Children 1988;55:343�7.

(A) Verdonschot EH, Angmar-Mansson E, ten Bosch JJ, Deery CH, Huysmans MC, Pitts NB, et al. Developments in caries diagnosis and their relationship to treatment decisions and the quality of care. Caries Res 1999;33:32�40.

Backer-Dirks O, Amerongen J van, Winkler KC. A reproducible method for caries evaluation. J Dent Res 1951;30:346�59.

Ekstrand K, Qvist V, Thylstrup A. Light microscope study of the effect of probing on the occlusal surfaces. Caries Res 1987;21:368�74.

Lunder N, von der Fehr FR. Approximal cavitation related to bitewing image and caries activity. Caries Res 1996;30:143�7.

Lussi A. The impact of including or excluding cavitated lesions when evaluating methods for the diagnosis of occlusal caries. Caries Res 1996;30:389�93.

Nielson A, Pitts NB. The clinical behavior of free smooth surface carious lesions monitored over two years in a group of Scottish children. Br Dent J 1991;171:313�8.

Pitts NB, Fyffe HE. The effect of varying diagnostic thresholds upon clinical caries data for a low prevalence group. J Dent Res 1988; 67:592�6.

Pitts NB, Rimmer PA. An in vivo comparison of radiographic and directly assessed clinical caries status of posterior approximal surfaces in primary and permanent teeth. Caries Res 1992;26:146�52.

Rimmer PA, Pitts NB. Effects of diagnostic threshold and overlapped approximal surfaces on reported caries status. Comm Dent Oral Epidemiol 1991;19:205�12.

Rimmer PA, Pitts NB. Temporary elective tooth separation as a diagnostic aid in general dental practice. Br Dent J 1990;169:87�92.

Scottish Intercollegiate Guidelines Network. SIGN guidelines: an introduction to SIGN methodology for the development of evidence-based clinical guidelines. Edinburgh: SIGN; 1999 (SIGN publication no. 39).

Abstracts Index