Selective carious tissue-removal strategies require specific considerations in selection of restorative materials. A tight marginal seal placed over hard dentin and sound enamel is essential. For selective removal of carious tissue with permanent restoration, bioactive materials, such as high-viscosity glass-ionomer cement (HV-GIC) or calcium silicates, may be preferred over caries-affected firm or leathery dentin to improve remineralization. HV-GICs have the best clinical evidence of caries-arresting effect and demonstrate sufficient longevity as long-term provisional restorations that can later be used in open or closed sandwich restorations. As with any material, oral health maintenance remains important for long-term survival of restorations.
Carious dentin can be left under provisional or permanent restoration without risk of caries progression.
Restoration marginal sealing must be tight and placed on intact enamel and dentin.
There is little evidence of any contemporary restorative material to be superior as temporary or permanent filling over carious dentin.
Cavity lining is not necessary for pulpal health or dentin remineralization.
With the advances in understanding the biofilm and improvements in the material and bonding technologies, the conventional concept of complete caries removal combined with extension into sound tissue for prevention was replaced with the minimal invasive selective or stepwise carious tissue-removal approach. The advent of selective removal of carious tissue has raised a question: which restorative materials should be used over carious tissue left under the restoration? Before discussing the performance of restorative materials, it is essential to realize that the approach to leave at least firm caries-affected dentin under the restoration is not necessarily new. Lesion in dentin does not have a distinct border between carious and healthy tissue, and judgment between “carious” and completely “healthy” dentin is clinically challenging if not impossible. Even with nonselective caries removal, some firm caries-affected dentin is left behind, as visual-tactile diagnosis of residual caries is very subjective method, and even the organic component of dentin may be altered in early phases of dentinal caries. The amount of caries-affected or caries-infected dentin left behind depends also on the method used for caries removal. It may thus be argued that clinicians have been practicing selective removal of carious tissue to reach firm or even leathery dentin even in the era when the nonselective “complete” caries removal was a must.
The choice of restorative material is affected by the carious tissue-removal approach: is the tooth permanently restored, or is stepwise carious tissue removal chosen to avoid pulp perforation and retain pulp vitality? If permanent restoration is placed immediately, clinical features such as the remaining coronal tooth tissue, the size of the restoration, occlusal forces, caries risk, and esthetics affect the material selection. This applies especially to carious tissue removal extended to firm dentin. If leathery dentin, which does not deform under pressure but is still relatively easily removed with an excavator, is left underneath, the clinician may also want to consider how this type of suboptimal tooth substrate may affect the behavior and longevity of the restoration. In addition, the location of the lesion and remaining carious tissue may significantly affect the long-term outcome of the restoration. It should be emphasized here that selective removal applies only to dentin facing the pulp: all carious enamel and soft-leathery dentin should always be removed from the cavity margins to ensure restoration placement to structurally and mechanically strong enamel ( Fig. 1 ) and to ensure a tight marginal seal.
Stepwise removal of carious dentin
Stepwise removal refers to two-step carious tissue removal, whereby in the first stage, selective removal is extended to soft dentin in the pulpal side and the cavity is sealed with a long-temporary restoration for 6 to 12 months. At control, the cavity is re-entered to remove the remaining soft carious tissue and place permanent restoration. The stepwise approach is particularly useful in deep lesions where the risk for pulp exposure with selective carious tissue removal is imminent.
The Use of Liners During Stepwise Removal of Carious Dentin
Cavity liners have been promoted at the pulpal walls in deep cavities to reduce hypersensitivity, kill bacteria, block the harmful components of the restorative materials from reaching the pulp, induce reactionary dentin formation, and/or promote remineralization. Calcium hydroxide (CaOH) especially has a long history as a lining material under restorations, including the early stepwise excavation studies. A systematic review concluded that CaOH is effective in reducing bacteria and promoting remineralization of remaining carious dentin. However, the review included 13 studies, of which 6 had no control group, 2 had CaOH in the control group, and 2 did not have CaOH in either experimental or control groups. More recent studies have indicated that the use of CaOH or any other contemporary liner material does not have a significant effect on the clinical or microbial outcome. Another systematic review and meta-analysis found no difference in pulpal response between CaOH and resin-modified glass-ionomer liners. Thus, the available evidence does not support the use of liners if good marginal seal of the cavity is achieved.
However, CaOH may have stronger antibacterial effects than other commonly used lining materials, and may reduce bacterial numbers much more effectively than only sealing the cavity. In the case of stepwise excavation close to the pulp, when remaining soft dentin is apparently heavily infected it may be advisable to use CaOH liner not only to reduce bacterial load but also to help prevent the accidental pulpal perforation during the re-entry, especially when tooth-colored materials are used as temporary filling.
Restorative Material During Stepwise Removal of Carious Dentin
Numerous materials have been used as temporary restorations in stepwise removal studies, and there is no apparent consensus of one material being superior to others. The requirements for the temporary restorative material are tight seal, sufficient longevity, and ease of removal during re-entry. Early studies with shorter follow-up time (usually 3 months or less) used zinc oxide-eugenol-based materials. However, the currently recommended 6- to 12-month follow-up time may require more durable material to ensure a marginal seal. Amalgam has a good longevity and is especially successful in patients at high risk of caries. The corrosion products seal the margin soon after placement, and antibacterial properties may protect against secondary carious lesions. Amalgam is also easy to discriminate from tooth tissue during removal, but poor aesthetics and increasing limitations for the use of mercury-containing materials and procedures (Minimata Convention on Mercury: mercuryconvention.org/ ) are limiting the use of dental amalgam worldwide. High-viscosity glass-ionomer cements (HV-GICs) have good biocompatibility, bond chemically to dental hard tissues, release fluoride, may protect against secondary caries, and have good longevity especially in class I and II cavities. Being less aesthetic than composite resin, it is easier to remove during re-entry, but still much more aesthetic than amalgam. Composite resins form a tight seal in cavity margins, but bonding to carious dentin, especially to soft carious dentin, is poor ( Figs. 2 and 3 ), and the alterations of mechanical properties in caries-affected dentin may subject dentin to cohesive damage caused by composite resin shrinkage stress (see Fig. 3 ). In addition, placing composite resin requires more time, and because of the tight bond to sound marginal dentin and enamel and good aesthetics, the loss of sound tissue may occur during re-entry. In conclusion, HV-GIC is the material of choice for a long-term provisional restoration in stepwise removal of carious dentin.
Selective removal of carious tissue
Selective removal of carious tissue may reach firm or leathery dentin, or can be left to soft (caries-infected) dentin. In very shallow lesions, removal to firm dentin may be necessary to reach the required thickness of the restorative material. In moderately deep cavities, relatively hard but leathery dentin can be left to the pulpal cavity side to avoid pulpal perforation. Clinical studies have indicated that vitality is retained with selective carious tissue removal in both primary and permanent teeth.
Remineralization of the partially demineralized carious dentin, reaching the original hardness and mechanical properties, is the ultimate goal of the restorative material research, but is not easy to achieve. The mechanical properties of mineralized dentin depend highly on intrafibrillar minerals, and true remineralization should aim at re-establishing intrafibrillar minerals. This, however, may be a difficult if not impossible task with the contemporary materials if the original nucleation site phosphoproteins have been lost at the space between the collagen molecules where the intrafibrillar mineral is sited (the gap area), and plain rehardening is still the best achievable goal.
Similar to stepwise removal of carious dentin already discussed, there are no clinical studies or consensus of the best restorative material over the caries-affected—leathery or firm—dentin. The advantages and disadvantages of contemporary direct restorative materials are now discussed in relation to caries.
Amalgam and composite resins are the most common “permanent” direct restoration materials globally. Recent meta-analysis indicated that amalgam restorations still have better longevity than composite resin restorations. However, it should be noted that most of the studies included in the meta-analysis were relatively old, excluding the advances in composite resin restorative materials and techniques. It should also be noted that traditionally nonselective caries removal to hard dentin has been practiced, and little is known of the effect on survival of leathery/firm caries-affected dentin left underneath on purpose during amalgam restorations.
Amalgam corrosion products seal the gap between the tooth tissue and restoration, and antibacterial properties of amalgam have been held responsible for the protection against secondary carious lesions. However, the corrosion products may also affect caries-affected dentin. Along with tin, zinc is consistently found under amalgam fillings in caries-affected dentin and in artificially demineralized dentin exposed to amalgam. Zinc-containing amalgams decrease the microleakage much faster than nonzinc amalgams, excess of zinc effectively inhibits collagen-degrading matrix metalloproteinases (MMPs) participating in the dentinal caries process and adhesive hybrid layer degradation, and Zn-containing amalgams promote remineralization and increase nanomechanical properties of subjacent caries-affected dentin. It is tempting to speculate that zinc anticollagenolytic/remineralizing effects may contribute to the longevity of amalgam restorations placed over caries-affected dentin.
As mentioned earlier, the immediate bond strength to caries-affected dentin is lower than to sound dentin and even lower with caries-infected dentin (see Fig. 2 ). The literature also indicates that etch-and-rinse adhesives may yield somewhat higher bond strength than self-etch materials to caries-affected dentin. In addition, dentin bond strength decreases with time regardless of the type of dentin (see Fig. 2 ), mainly because of collagenolytic enzymes degrading the adhesive hybrid layer collagen and hydrolytic degradation of the resin. Clinically available methods to achieve collagenolytic enzyme inhibition and improve bond-strength durability include cavity pretreatment with chlorhexidine and the use of quaternary ammonium methacrylate (QAM)-containing materials. QAMs, such as 12-methacryloyloxydodecylpyridinium bromide (MDPB), present in some clinically available adhesives, are polymerizable antimicrobial compounds and the most versatile, most studied, and perhaps most promising adhesive resin components with antibacterial properties. In addition, they also inhibit dentin MMPs, which has been indicated as a potential reason for prolonged durability of dentin bonding in vitro . Despite very promising in vitro and in vivo experimental results, the evidence of improvement in clinical conditions is still lacking. The use of noncarious cervical lesions instead of carious lesions and the still short follow-up times may have contributed to the results of the clinical studies. Therefore clinically, bonding to caries-affected and especially caries-infected dentin can only be recommended if the area is very limited and the peripheral seal of the cavity is placed on sound dentin and enamel to ensure a durable bonding.
Cariogenic bacteria, such as Streptococcus mutans , readily accumulate on the surface of composite resin, and improving the antimicrobial properties of the restorative or adhesive materials is an attractive alternative both to reduce the adverse effects of bacteria remaining under the restoration and to reduce secondary caries. Demineralization is not the only mechanism whereby bacteria may accelerate secondary caries at the interface. Cariogenic bacteria possess esterase activity in levels that can degrade resin composites and adhesives, and degradation products may further increase the expression of the esterase and accelerate the biodegradation of the restoration components. Several attempts have been made to introduce antimicrobial biofilm-modulating properties of composite restorative materials, but there is very little evidence of clinical benefits of these materials.
Fluoride-releasing materials—GICs, resin-modified GICs (RM-GICs), and compomers, are promoted as cariostatic restorative materials and as an alternative to composite or amalgam restorations. Although the fluoride release of RM-GIC and compomer is significantly lower than of GIC ( Fig. 4 ) and has not been shown to have any significant effect, there is emerging support for the therapeutic effect of GIC against caries, especially for HV-GIC. Meta-analyses indicate a clinical caries-preventive effect of GIC, but because of the lack of randomized controlled studies the results are limited to comparison with amalgam and single-surface fillings in permanent teeth. Unlike contemporary adhesives, which have a significantly lower bond strength to caries-affected dentin than sound dentin regardless of the adhesive system used (see Fig. 2 ), HV-GICs have similar bond strength to both normal and caries-affected dentin. El-Deeb and Mobarak concluded that HV-GICs can be recommended to restore cavities after selective removal of carious dentin.
Postirradiation xerostomic patients present very high caries rates, and provide an ideal population for investigating the effects of restorative material in a reduced time frame. Several studies have demonstrated good survival rates for HV-GICs in these patients ( Fig. 5 ). More notably, both low-viscosity GIC (LV-GIC) and HV-GIC restorations completely prevented secondary caries even in patients with low compliance to fluoride use or when the restoration was completely lost, and the deteriorated restorations could also be turned into sandwich restorations.