Evidence based management of dental caries; a review of the repair potential of the pulp-dentine R.J.C.Wilding. BDS, Dip Pros. M.Dent. Ph.D Home Page

"If seven maids with seven mops, swept it for half a year, do you suppose" the walrus said, "That they could get it clear?"
"I doubt it" said the Carpenter, and shed a bitter tear. Lewis Carol

The futility of the maids and their mops made the Carpenter weep. The futility of the dentists and their drills is equally sad. More than ten years ago, Louw (1982) calculated that it would take 470 dentists, working all year long, to restore the carious lesions in 12-year old 'coloured' children in South Africa. It goes without saying that this massive damage could have been prevented; clearly, the way forward is to promote dental health and apply preventive measures. But in developing nations, and increasingly in some population sectors of developed nations, we still have to deal with the legacy of neglect. In 1996 the Commonwealth Dental Association and WHO sponsored a three day workshop on equity in oral health (CDA/WHO 1996). On of the many challenges addressed at this workshop was how to provide funding for the resources required. All our collective ingenuity will be required to develop appropriate strategies to manage this problem. The purpose of this review is to contribute to finding a solution to this challenge, by reviewing the evidence for managing caries with low intervention, low cost procedures.

The active role players in caries

Dental caries is caused by interaction between oral bacteria, their access to fermentable carbohydrates and vulnerable parts of the tooth. The now classic graph which bears Stephan's name, shows the rapid drop in plaque pH after a glucose rinse (Stephan and Miller, 1943; Fig 1). The drop in pH is the result of fermentation of carbohydrates by some plaque bacteria. The gradual return of the pH is the result of buffers present in plaque and saliva. Provided the pH does not drop below 5.3 the enamel remains intact, but below this critical level, crystals of apatite dissolve. Fortunately both plaque and saliva are saturated with calcium and phosphate ions, so that if the pH returns fairly rapidly above the 5.3 level, ions will go back into the enamel and recrystallise. This process of remineralisation takes longer in an acid environment, but is rapid if the fluid next to the enamel is neutral or even alkali. Caries is therefore uncommon in those parts of the mouth near the outflow of salivary glands, like the lower incisors, where the teeth are constantly bathed with the buffers and concentrated calcium ions of saliva. If the total outflow of saliva can be increased, there is a greater chance of protection of all the teeth in the arch. Some foods, like cheese, stimulate the flow of saliva. Jenkins (1970) observed that cheese after a meal was not only a pleasant way to end it, but also a particularly good caries inhibitor. Sugar is also a good stimulator of saliva but of course also provides most favoured nutrients for plaque bacteria. However, sugar substitutes, while of no use to bacteria, are also effective saliva stimulants, as is the mere act of chewing (Hector, 1985), so chewing a gum which is artificially sweetened makes biological sense, and indeed has a role in arresting caries (Kleinberg, 1985; Makinen et al, 1995).
If the flow of saliva is reduced as it is every night, the oral environment is particularly susceptible to plaque acids. Sweets at bedtime, thus have a far greater impact on acid production during the night, than they do during the day. Stephan's curve can also be used to demonstrate the effect of tooth brushing on the pH reached by plaque after a second glucose mouth rinse (Fig 1).

Figure 1. A plot of the pH of plaque before and after a glucose mouth rinse. At (1), the pH drops due to acid produced by bacteria fermenting the sucrose. At a pH of 5.3 enamel begins to dissolve. At (2), the PH rises due to the buffering action of plaque and saliva. At (3), one side of the arch is brushed and the pH rises to neutral value of 7.0. At (4), a second glucose rinse causes the plaque pH on the unbrushed side to again drop below 5.3. The brushed side also drops but not below the critical pH level (After Stephan and Miller, 1943)

Reducing the plaque mass raises the resting pH and reduces the downward dip of the curve after a second glucose rinse. The mass of plaque bacteria can also be reduced with antibacterial mouthwashes such as chlorhexidene (Joyston-Bechal et al, 1992).

In summary there is an ebb and flow of minerals into and out of the enamel. Caries occurs when the process of remineralisation is slower than the process of demineralisation and there is a nett loss of mineral into the environment. It can be prevented by restricting the intake of dietary sugars, and removing plaque; at least these are the methods promoted by dentists in their own families (McDonald, Cowell and Sheiham, 1981). These simple habits seem to have worked, as dentists' children have less caries than the general population (Ainamo and Holmberg, 1974).

Early enamel caries

Enamel is almost entirely mineral by weight (96%) but only 87% mineral by volume. Thus 13 % of the space in enamel is water and soluble and insoluble proteins (Le Geros, 1991). The organic and water component of enamel allow diffusion of ions from plaque and saliva into and out of enamel. The mineral part of enamel consists mostly of varieties of biological apatites. An early enamel lesion seen under polarised light reveals four distinct zones of mineralisation (Fig 2).

Figure 2. The early enamel lesion consists of four zones of alternating levels of mineralisation. It illustrates the dynamic nature of the caries process. The surface zone blocks the passage of calcium ions into the body of the lesion and may have to be removed to allow the lesion to become arrested (After Kidd & Joyston-Bechal, 1987).

The outer surface zone is well mineralised by replacement ions from plaque and saliva. But the body of the lesion is poorly mineralised. Deeper to the body of the lesion, a darker zone represents some remineralisation, while the deepest zone, is yet again demineralised (Kidd and Joyston-Bechal, 1987). These zones illustrate the dynamic series of events which are occurring in the early lesion. Caries is not simply a process of continued demineralisation.

Dentine Caries

Although caries of enamel is clearly a dynamic process, it is not a vital process in the sense that living cellular reactions occur. The pulp and its dentine is a fully vital tissue capable of defending itself. The essential relationship between pulp and dentine promotes the term pulp-dentine as a structural and functional biological unit.

The caries process in dentine involves the demineralisation of the mineral component and breakdown of the organic component of collagen fibres. The caries process in dentine is approximately twice as rapid in enamel. Advanced carious lesions in dentine consist of two distinct layers having different microscopic and chemical structures (Daculsi et al, 1987). The outer layer is heavily infected by bacteria which are mainly located in the tubule spaces. The collagen fibres are denatured and the organic matric is not being remineralised. The inner layer is scarcely infected, but affected by plaque acid. It still contains high concentrations of mineral salts and can be remineralised (Fig 3).

Microbiology of dentine caries

Although a large number of organisms have been isolated from dental caries a few genera are commonly found and predominate. The most frequently isolated are the members of the Streptococci sp. and in particular S. mutans, in occlusal and smooth surface caries. Actinomyces sp. are the dominant genus in root surface caries (Roth and Calmes 1981). Deep dentinal caries shows a predominance of Lactobacillus organisms with several other gram positive rods and filaments.

Kidd, Joyston-Bechal and Beighton (1993) took samples of carious dentine during cavity preparation, and cultured the samples so as to count the number of bacteria. As the samples were taken, the dentine was assessed as either soft, medium or hard, wet or dry and pale or dark. The number of bacteria recovered diminished significantly as the caries became dryer and harder and the cavity became deeper (Fig 3).

Figure 3. Dentine caries comprises two main layers. In the outer layer, the dentine is heavily infected with bacteria. Both organic matrix and mineral have been lost and the dentine is beyond repair. In the deeper layer, the dentine has been affected by plaque acids and demineralised. The number of colony forming units (CFU) of bacteria decreases (about 100 times) as cavity preparation proceeds into affected dentine. The damage in this layer is reversible if bacterial metabolism can be halted. A barrier of translucent (well mineralised) dentine may be formed ahead of the advancing lesion. Reactionary (secondary ) dentine forms to protect the pulp from acid irritation (After Kidd & Joyston-Bechal, 1987).

The studies on marginal leakage were the first indications that secondary caries was not always caused by incomplete cavity preparation. According to Edwardsson (1987) the microflora left beneath restorations usually do not grow, although they may survive for several months. Their viability is mainly influenced by the amount of carious dentine sealed in and the degree to which they are sealed off from the oral cavity.

Conservative management of deep caries

A traditional test of adequate caries removal during cavity preparation, is to hear the ring of a sharp probe on a hard dentine floor. Massler (1962) promoted the idea that this was unnecessarily destructive, and could often lead to exposure of the pulp in deep cavities. He maintained that soft dentine was not necessarily infected, but could be merely affected by plaque acids. Such an affect was reversible and hence this dentine could remineralise if given a suitable environment. He proposed a more conservative cavity preparation, which required the removal of infected dentine only, leaving behind soft but not infected, so called affected dentine. Affected dentine could be remineralised if the acid production was halted. Although Massler's ideas were supported by little experimental data, the idea of leaving behind soft dentine, and revisiting the tooth 6 weeks later, became an acceptable practice. It was more recently observed that the 6 weeks period was too short to allow a significant amount of secondary dentine to form, but it did allow an assessment of the pulp status to be made (Schroder, 1985). Massler(1967) later argued, with more data, that an inflamed pulp was not a dead one, and could recover if the irritation was removed. He promoted the idea that the pulp-dentine had great reparative power and that this should not be ignored in managing caries.

Arrested Caries,a>

Arrested caries in dentine is clinically defined by hardness of the dentine surface and a yellow to dark brown colour. Arrested carious lesions are found most commonly on lingual and labial aspects of teeth and less commonly interproximally. In caries which has become arrested, the dentinal tubules in the area between the soft and hard dentine have been shown to be obstructed by large crystals. This process appears to occur in a number of stages (Fig 4).

Figure 4. Steps in the formation of an arrested lesion in dentine a) The dentine tubule contains a high concentration of dissolved mineral salts. b) If bacterial acid production is reduced, and the pH increases, the salts precipitate into large crystals of tricalcium phosphate which temporally block the tubule. c) If further bacterial activity is suppressed, the odontoblast secretes collagen and calcium salts. Crystals of hydroxyapatite then form and block the tubule more effectively (After Daculsi et al, 1987)

First Stage; The acids produced by advancing bacteria dissolve the mineral in the surrounding intertubular dentine. The tubule fluid becomes saturated with calcium magnesium and phosphate ions. The lesion progresses unless the level of metabolic activity of the bacterial is reduced. If less acid is produced then the second stage can occur.

Second Stage; The saturated solution precipitates, producing large crystals of tricalcium phosphate. These crystals are comparatively soluble but nevertheless block the tubule.

Third stage; The odontoblast process, protected by the large crystals blocking the tubule, secretes collagen into the dentin tubule. Small plate-like crystals of hydroxyapatite accumulate, which are less soluble than tricalcium phosphate and therefore block the tubule more effectively. At the same time crystal growth occurs in the intertubular dentine.

Strategies for promoting the arrest of caries

Fluoride

Fluoride ions increase the resistance of the hydroxyapatite in enamel and dentine to dissolution by plaque acids. The source of fluoride ions, to carry out this protective role, has mostly been via toothpastes. In fact it is fluoridated toothpaste which has been thought responsible for the reduction in caries which has occurred in many developed countries (Sheiham 1994). In developing countries, where the cost of toothpaste, places it beyond the reach of many, water fluoridation is an economical and effective alternative. It is becoming apparent, that the most potent effect of water fluoridation, is not so much in preventing new lesions from appearing, but on remineralising existing carious lesions, and thus slowing down, or even arresting the carious process (Backer-Dirks, 1961; Lawrence, Sheiham and Benn).

The greatest benefit from fluoride is achieved when there is a constant low level available for remineralisation. In vitro studies indicate that provided fluoride is in concentrations above 2 ppm, substantial remineralisation occurs on artificially created lesions ( Arends et al, 1989). The levels required by dentine are however much greater than enamel. While enamel remineralises well with less than 5 ppm of fluoride, dentine requires over 100 ppm. Fluoride releasing restorative materials such as the glass ionomer cements have been show to release fluoride for periods up to 6 months. Glass ionomer cements used to retain orthodontic brackets cause significant increases in the fluoride concentrations found in adjacent plaque (Hallgren, Oliveby and Twetman, 1993).

Gluterdialdehyde

A 2 minute daily application of GDA (gluterdialdehyde) has been shown to reduce mineral loss in dentine caries, presumably as a result of collagen fixation, reduced diffusion of ions out of the lesion and its antibacterial action (Dijkman, De Vries and Arends, 1992). These results indicate that GDA may be able to contribute to arresting dentine caries and could be used in institutions such as schools, where GDA could be applied under daily supervision.

Sugar substitutes

Chewing gums containing sugar substitutes have been found to be effective in retarding or arresting rampant dentine caries. Makinen et al, (1995) used xylitol and sorbitol containing gums in a supervised gum chewing trial over 40 months on 510, 6-year olds and 1277, 10-year old subjects. Rehardening of dentine caries was observed most frequently in groups chewing gum containing xylitol. The authors suggest it is the increased saliva flow resulting from the sweet tasting gum which tends to arrest existing carious lesions. Supervised activities are costly in resources, but it may be possible to interest the private sector in subsidising the manufacture of sugar free gum if its use could be endorsed by the profession as a healthy habit.

Low intervention procedures for arresting caries.

Making cavities self cleaning.

The removal of plaque from a carious lesion causes a shift in the balance of factors causing demineralisation. If existing cavities can be made accessible to cleaning, the carious process may be arrested. The potential for minimal enamel removal rendering a cavity cleanable is probably greatest in smooth surface caries. This technique which Massler(1962) referred to as saucerising was a real alternative for shallow root surface caries before the advent of dentine bonding resins. Its value is clearly dependent on motivating the patient to clean the cavity especially thoroughly.

Sealing in caries.

The removal of infected dentine would appear to be the minimal procedure to insure long term success with a restoration. However the unthinkable was done, and has survived well over ten years now. Mertz-Fairhurst and her colleagues (1995) at the Medical College of Georgia, sealed in caries using composite materials. The study involved over 120 patients. There were three restorative methods used. The first method involved making an ultraconservative preparation in which no caries was removed but the enamel margins were bevelled. These cavities were filled with a posterior composite resin (chemically cured) and the entire surface and surrounding fissures covered with a tinted sealant. After 9 years the restorations were assessed radiographically and clinically. The cumulative failure rate, mostly in terms of marginal breakdown, was 16% for the composite restorations, compared with 17% for the amalgam controls. The authors claim that caries can be arrested by placing sealed composite restorations. However these restorations have not been opened up and examined, so the degree of remineralisation cannot be accurately assessed. The radiographic evidence is that the lesions have not progressed.

Atraumatic Restorative Technique

Following the same philosophy as the Mertz-Fairhurst team, a technique has been described which claims to avoid even rotary instrumentation in the cavity preparation. Frencken et al, (1994) report on a trial that took place in rural Thailand, in which dental cavities were prepared using hand instrumentation to remove undermined enamel and soft dentine. No anaesthesia was required. The cavities were restored with a glass ionomer cement as both filling material and sealant. Mobile dental units were used to provide a traditional amalgam treatment of caries in another village. After one year 79% of ART fillings in deciduous teeth and 93% in permanent teeth were successful. The authors found that two surface restorations were less successful. This technique pushes materials to the edge of their properties and challenges the grounds of many clinical standards. The absence of rotary instruments places a major constraint on the preparation of margins which can be made accessible and free of loose enamel prisms. The use of glass ionomer has the benefit of fluoride release and excellent bonding to both dentine and enamel, but does not have the resistance to wear and fracture of filled composites.

All three of these rather unusual approaches to dental caries are of direct interest where treatment resources are limited. Firstly they could all be performed without anaesthesia. If dentine is sensitive it should not be removed. The sensitivity of dentine to being cut depends on the transmission to the pulp, of fluid vibrations within the tubule. If the tubules are patent the processes described by Daculsi et al, (1979) should occur in which a mineral barrier and a collagen plug arrest further caries activity. Secondly they could be performed by operators with less than a full dental training. A thorough knowledge of the biology of the pulp-dentine is not demanded if the only preparation were to bevel enamel margins.

Caries risk assessment

Caries is unlikely to become arrested in individuals or groups who have a high caries risk (Moss and Zero 1995). It is not easy to identify who such groups or individuals are, but there are some obvious contraindications to using low intervention methods to arrest caries. The first must be the level of health awareness of the individual and the level of motivation for applying effective oral hygiene practices and sensible eating habits. Minimal procedures for treating caries are not for the unmotivated. Unfortunately it is often the poor, and poorly educated, who are targeted as the recipients of low intervention techniques.

A known caries risk is radiotherapy, after which the rate of salivary secretion is diminished and many patients experience new caries lesions. Such high risk patients can be protected during the period of radiotherapy and salivary gland damage, by using a twice daily mouth rinse of chlorhexidine and fluoride (Joyston-Bechal et al, 1992). However this would not be a group for whom low intervention procedures for treating existing caries would be indicated. It is necessary, therefore, to ensure that public health programmes match the management of the disease to the risk.

Conclusions

The inherent repair capacity of the pulp-dentine should be recognised. Patient education is the first step towards establishing an environment which supports remineralisation. As Kidd and Joyston-Bechal (1987) urge, "It is the dentists responsibility to explain to the patient how they may arrest disease in their mouths". The selection of patients for whom limited clinical procedures are appropriate, is difficult. It would be hard, at present, to defend limited procedures, against the accusation that they represent second class care for the underprivileged. Low intervention procedures will first have to be fully investigated, to determine whether in fact they could be beneficial and low cost. Continued basic research into the dynamics of arresting caries will clearly also contribute to evidence based strategies. The need to devise solutions which are socially acceptable, and scientifically valid is a challenge which deserves urgent attention. However the priority remains to put in place health promotion, preventive measures and appropriate water fluoridation for everyone. We will otherwise be sweeping up sand, with no end in sight.

References

Ainamo, J and Holmberg, SM. (1974) Oral health of dentists, children. Scand J. dent. Res. 82, 547-551.

Backer-Dirks, O. (1961) Longitudinal dental careis study in children 9-15 years of age. Archs. oral Biol. 6. 94-108.

Arends, J, Christoffersen, J, Ruben, J & Jongebloed, WL (1989) Remineralisation of bovine dentine in vitro. The influence of the F content in solution on mineral distribution. Caries Res. 23. 309-14.

CDA/WHO. Proceedings of a workshop on Promoting Equity in Oral Health, Cape Town 1-3 April 1996. Ed N.G. Myburgh. University of the Western Cape.

Daculsi G, Le Geros RZ, Jean, A & Kerebel, B (1987) Possible physico-chemical processes in human dentine caries. J Dent Res. 66, 1356-9.

Dijkman, GE, de Vries, J & Arends, J (1992) Effect of gluterdialdehyde on secondary caries in situ. Caries Res. 26. 293-8.

Edwardsson, S (1987) Bacteriology of dentin caries. Dentine and dentine reactions in the oral cavity. Ed. Thylstrup A, Leach SA & Qvist V. IRL PRess.

Frencken, JE, Songpaisan, Y, Phantumvanit, P & Pilot, T (1994) An atrauamatic treatment (ART) technique; evaluation after one year. Int Dent J. 44. 460-4.

Hallgren, A, Oliveby, A & Twetman, S (1993) Fluoride concentration in plaque adjacent to orthodontic appliances retained with glass ionomer cement. Caries Res. 27. 51-54.

Hector, MP (1985) The masticatory salivary reflex. In: Current Topics in Oral Biology. Ed Lisney and Matthews. University of Bristol.

Holmen, L, Thylstrup, A & Artun, J (1987) Surface changes during the arrest of active enamel carious lesions in vivo. Acta Odontol Scand 45. 383-390.

Jenkins, GN (1970) The physiology of the mouth. Blackwell.

Joyston-Bechal, S, Hayes, HK, Davenport, ES & Hardie, JM (1992) Caries Incidence, mutans streptococci and Lactobacilli in irradiated patients during a 12-month preventive programme using chlorhexidene and fluoride. Caries Res. 26. 384-390.

Kidd, EAM & Joyston-Bechal, S (1987) Essentials of Dental Caries. Wright. Bristol.

Kidd, EAM, Joyston-Bechal, S & Beighton, D (1993) Microbiological validation of assessments of caries activity during cavity preparation. Caries Res. 27. 402-408.

Kleinberg, I (1985) Oral effects of sugars and sweeteners. Int Dent J. 35. 180-189.

Lawrence, HP, Sheiham, A and Benn, DK. Caries progression in a population of Brazilian school children in fluoridated and fluoride-deficient areas. Comm. Dent. Oral Epid. Submitted for publication.

Le Geros, RZ (1991) Calcium Phosphates in Oral Biology and Medicine. Karger.

Louw, NP (1982) Essential needs of the Coloured Population in the Cape Peninsular PhD. Thesis, University of Stellenbosch.

Massler, M (1962) Control of caries; a new concept. New Zealand Dent J. 58. 69-73.

Massler, M (1967) Changing concepts in the treatment of carious lesions. Brit Dent J. 123. 547-8.

Makinen, KK, Makinen, PL, Pape, HR, Allen, P, Bennett, CA, Isokangas, PJ & Isotupa, KP (1995) Stabilisation of rampant caries: polyol gums and arrest of dentine caries in two long-term cohort studies in young subjects. Int Dent J. 45 (Suppl 1) 93-107.

McDonald, SP, Cowell,CR and Sheiham, A. (1981) Methods of preventing dental caries used by dentists for their own children. Brit.dent J.151, 118-121.

Mertz-Fairhurst, EJ, Adiar, SM, Sams, DR, Curtis, JW Jr, Ergle, JW, Hawkins, KI et al. (1995) Cariostatic and ultraconservative sealed restorations;Nine year results among children and adults. ASDC-J Dent Child 62. 97-107.

Mor, BM & Rodda, JC (1983) In vitro-mineralisation of artificial caries -like lesions with milk. NZ. Dent J 79, (10-15.

Moss, ME & Zero, DT (1995) An overview of caries risk assessment, and its potential utility. J Dent Educat. 59 932 -40.

Robinson, C, Hallsworth, AS, Shore, RC & Kirkham, J (1990) Effect of surface zone demineralisation on the access of mineral ions into subsurface carious lesions of human enamel. Caries Res. 24. 226-230.

Robinson, C, Kirkham, J, Bavenstock, AC & Shore, RC (1992) A flexible and rapid pH cycling procedure for investigations into the remineralisation and demineralisation behaviour of human enamel. Caries Res 26. 14-17.

Roth, GI & Calmes, R (1981) Oral Biology . C.V. Mosby

Schroder, U. (1985) Effects of calcium hydroxide-containing pulp capping agents on pulp cell migration proliferation and differentiation. J Dent Res 64 (Spec Iss) 541-548

Sheiham, A (1994) Oral health policy and prevention. In: Preventing Oral disease. Ed Murray.

Stephan, RM & Miller, BF (1943) A quantitative method for evaluating physical and chemical agents which modify production of acids in bacterial plaques on human teeth. J. Dent Res 22. 45-51.

This page was prepared by Robin Wilding and last updated on November 16th 1999. Write to RobWilding@eclipse.co.uk with comments or visit Moorland Dentistry