Minimally invasive long-term management of direct restorations: the ‘5 rs’

What is a ‘failing’ restoration?

A failing restoration can be described as one that has suffered biomechanical defect or damage resulting in immediate or subsequent detrimental clinical consequences to the patient. This may affect the restoration alone (eg bulk fracture, staining etc), the supporting tooth structure (eg fractured cusps, new caries at the tooth-restoration surface (CARS) etc) or, more commonly, both, affecting the collective tooth-restoration complex. Such failure can present as obvious fractures of this complex, possibly detectable active caries associated with restoration/sealant surface (CARS, previously described as secondary or recurrent caries) or can be more subtle, such as marginal discoloration of an anterior aesthetic resin composite restoration or marginal ditching of a posterior restoration.

A number of clinical indices have been developed to help classify the extent of restoration failure, including the Ryge & Snyder Index (1973), useful for research analysis as well as clinical management (summarized in Table 1).1 A more recent classification by Hickel et al (2010) defined the clinical criteria for the evaluation of direct and indirect restorations.2 This index includes three separate groups:

Each restoration is evaluated against these criteria and given a score out of five, depending on the clinical findings. This classification has been proposed as a tool to evaluate and standardize new restorative materials, a method to determine if restorations require repair or replacement and a quality assessment tool for reviewing dental restorations. This classification has been shown to be more sensitive at determining differences between restorations than older classifications.2 There are a number of challenges, which include the universal uptake of the new classification system and how the scoring equates to long-term survival in longitudinal studies. However, with a degree of operator calibration, the new clinical criteria and scoring system is a flexible method which may reduce the risk of clinically unnecessary restoration replacement.2

A collective summary of those criteria that can be used to assess the extent of tooth-restoration complex failure is presented in Table 2. It must be stressed that this final diagnosis must include patient factors and the clinical/biological impact of the failure. A ‘failing’ restoration in a patient with a low caries risk and low aesthetic demands is likely to be managed very differently from a patient with a high caries risk and high aesthetic demands.2 Indeed, a restoration may present with a detectable defect that results in no medium- to long-term detrimental consequence to the patient. This potentially larger technical failure may be managed differently compared to a more minimal defect with a clear detrimental biological consequence (eg active CARS). Consideration of all factors is imperative before making the decision to intervene operatively. This decision-making process must take into account the patients' expectations and their attitude to taking responsibility for maintaining their personal oral health. Management options and discussions should be documented clearly and this process may be further enhanced by the use of consented clinical photographic records.

Why do restorations fail?

Even when placed in ‘ideal’ circumstances, the majority of dental restorations may be expected to deteriorate gradually and ultimately fail at some future point as a result of the biological environment and functional forces to which they are subjected. Even if the ‘ideal’ principles of tooth preparation/cavity design, material choice, handling and placement are followed judiciously, it is impossible to recreate nature's design for an indefinite period.

The aetiology of failure of the tooth-restoration complex can be divided into mechanical or biological, with the most common biological cause cited as CARS (summarized in Table 3).3 Manhart et al reviewed the clinical performance of direct restorations in the permanent dentition and determined their longevity and annual failure rates. The mean annual failure rates for posterior stress-bearing amalgam restorations were 3.0% and 2.2% for equivalent resin composites. The reasons for failure in this study included secondary caries, fracture, marginal deficiencies, wear and post-operative sensitivity.4 The management of failing restorations is associated with a significant amount of clinical time and replacement of amalgam and resin composite restorations constitutes approximately a quarter of all the conservative dentistry carried out in general dental practice in the UK.5^,6^,7

Factors affecting tooth-restoration outcome

In order to minimize failure and promote clinical longevity, it is essential to appreciate the three fundamental factors which influence the long-term success of direct (or indirect) restorations:

The operator effect

The subjective diagnosis of restorative failure leads to significant inter-operator variability in its assessment.^13,14,15 Complexities arise in the infinite variety of presentations of a failing tooth-restoration complex, coupled with patient factors. Differences in operator gender and system of remuneration have been reported,13 McAndrew et al demonstrated that training clinicians in the assessment and diagnosis of restoration failure increases the accuracy and predictability of decision-making and reduced the frequency of restoration replacement compared to a non-trained control group.14^,15

Why repair restorations?

Restoration repair aims to promote tooth-restoration longevity and is less likely to result in detrimental, iatrogenic damage to hard tissues. The pulp, chairside clinical time is reduced and the need for local anaesthesia is often obviated.5^,8 Replacing restorations tends to drive the restorative cycle towards failure by removing excessive and unnecessary quantities of natural tooth structure, further weakening the final tooth-restoration complex.16^,17Table 4 lists the possible adverse consequences of regular total restoration replacement.

The ‘5 Rs’

Minimally invasive non-operative management options for the failing tooth-restoration complex may be divided into five categories (Table 5):

It is imperative to understand that all the listed management options will require appropriate review intervals based on the individual patient's risk factors.

Contra-indications for repairing restorations

The factors which influence the decision to replace restorations include the quantity of the tooth-restoration complex that is lost and/or presence of significant levels of active caries undermining most of the restoration, especially if previous repair attempts have failed. Care also has to be taken if the existing material type cannot be ascertained, as the outcome of the technical repair procedure may be unpredictable.5 It is assumed that patients with a high caries risk, or those who were infrequent attenders, were less suitable for MI restorative techniques.5 It can be argued equally, however, that these patients would actually benefit more from MI restoration resealing, refurbishment and repair as opposed to multiple replacements.

MI techniques to manage failing direct restorations

Reseal

The application of a flowable resin-based material to seal a non-carious marginal gap defect in an amalgam restoration (Figure 5) has been shown to increase the longevity of the restoration.24 Prior to its application, the tooth-amalgam interface can be modified using a variety of techniques, including a rotary bur or air-abrasion, etching with phosphoric acid and application of a dentine-bonding agent.25

Repair

Amalgam restorations can be repaired with amalgam, resin composite or as a temporary/provisional measure with glass ionomer or resin-modified glass ionomer cements.

Amalgam to amalgam repair

Reports on shear bond strength between a newly placed and existing amalgam vary, with some reporting high values26 and others significantly lower.27^,28^,29 A number of methods have been proposed to increase the bond strength between existing and new amalgam, including macro-mechanical retentive undercuts, slots, pins, grooves, amalgam bonding28 (with a limited evidence base for successful use) and dentine adhesives.30 It is strongly recommended, whichever repair method is used, that the surface of the old amalgam should be roughened prior to the addition of new amalgam with the use of a rotary coarse diamond grit bur or 27–50 µm aluminum oxide air-abrasion, if available.31 The use of macro-mechanical features, such as retentive locks in the existing amalgam, should be considered to support the repair (Figure 6).

Resin – amalgam repair

Resin composite repair of amalgam restorations is considered to be a suitable method providing the appropriate protocol is followed.32 Improved bond strengths have been reported between resin composite and amalgam, with more predictable bonding and less microleakage between the repair-tooth interface and repair-restoration interface compared with an amalgam repair.33^,34 However, microleakage at both the restoration and tooth interface compared to that found in bonded amalgams was not eliminated completely. Surface conditioning of the amalgam prior to application of resin composite can be achieved using mechanical, micro-mechanical or chemical means. Micro-mechanical and chemical retention have been shown to be useful in situations where macro-mechanical retention alone is difficult to achieve, such as a fractured cusp on a premolar tooth (Figure 7).35

Various techniques to roughen the amalgam surface and increase micro-mechanical retention prior to bonding composite have been suggested.36 An appropriate method for bonding resin composite to amalgam is using air abrasion on the amalgam surface with silica oxide particles (CoJet-Sand, 30 µm SiO₂ 3M ESPE AG, Seefeld Germany) followed by a silane-coupling agent, a bonding agent and a resin composite.34 The silica has a tribochemical effect at the amalgam surface and, on application of silane, forms a polysiloxane network, which reacts with resin composite methacrylate groups.35

The type of adhesive has also been shown to influence the microleakage with etch and rinse adhesives (Types 1 and 2; 4th and 5th generation) performing better than self-etching primer adhesives (Types 3 and 4; 6th and 7th generation).37

GIC/RMGIC – amalgam repair

The properties of GIC make it an unsuitable long-term restorative material for load-bearing restorations in posterior teeth, even with newer materials such as Chemfil Rock (Dentsply).38 RMGIC is also unsuitable as a definitive restorative material owing to its increased water absorption and compromised aesthetics.39 Thus, GIC and RMGIC are only suitable for short- or medium-term repairs of amalgam restorations.

Resin composite

In contrast to dental amalgam, resin composite restorations present a more complicated substrate for repair. There is variation in the chemical composition of resin composites' matrix with additional disparities in inorganic filler quantity and type. Resin composites are based on two broad types of resin matrix: dimethacrylate-based or silorane-based.40

Ageing also has a significant effect on the bond strength between new to old resin composite restorations, decreasing with time.41^,42 Newly placed resin composites gain cohesive strength owing to the presence of uncured resin monomer at the restoration surface due to air inhibition (the oxygen-inhibited layer). This degrades rapidly on the exposed surfaces in the oral environment due to water adsorption and abrasion and thus prevents further incremental addition of resin composite.43

Dentists have been shown to be more likely to intervene operatively when a resin composite restoration is present in contrast to an amalgam.44 This may be due to a perceived increased longevity of amalgam restorations in general and the lack of correlation between marginal breakdown and the presence of secondary caries adjacent to amalgam restorations.

Review

In parallel with amalgam, if there is no clinical advantage to treating the composite restoration operatively, it should be reviewed (Figure 8).

Refurbishment

Resin composite restorations can be refurbished where defects are adjustable without damaging the underlying tooth structure.2 Polishing, improvement in surface roughness and anatomical form of Class I and II resin composite restorations have been shown to maintain an improvement at 3 years.22 The use of intra-oral air abrasion devices is a useful technique to refurbish discoloured resin composite restorations by removing superficial extrinsic staining. Lower abrasive powders, such as ClinPro Prophy Powder™ (3M EPSE, Seefeld, Germany), are recommended to prevent excessive abrasion of restorative material (Figure 9).45

Repair

Methods used to condition existing resin composite surfaces prior to the addition of new restorative material include surface roughening with a diamond bur, air-abrasion with aluminum oxide, bioactive glass particles, silica oxide (tribochemical particles), etching with 37% phosphoric acid and etching with 9.6% hydrofluoric acid.47

The literature identifies that one of the most important factors when repairing resin composites is the production of a strong cohesive bond between existing and new resin composites by using compatible materials of the same type, where possible, so producing a more predictable bond.45 As the majority of resin composites consist of similar properties, it is pragmatic to conclude that they can be repaired with any resin composite material. It is clear that operator technique is the most important variable in this repair process and that following a sound MI protocol is integral to producing successful outcomes.

Loomans et al investigated an optimal repair technique for all resin composites.49 The resin composites tested, notably nano-hybrid materials, micro-hybrid composite and hybrid composite containing quartz, exhibited repair bond strengths which varied, depending on the repair method used. A universal repair technique was difficult to achieve but, where the composition of the composite is unknown, a standard protocol would be roughening of the composite with a diamond bur then etching with 37% phosphoric acid for 20s or using air abrasion with silica oxide particles for 20s at a pressure of 1.5 bar at a distance of 20 mm. A combination of an organo-silane (which should be applied for 15s and gently air-dried) and bonding agent (adhesive resin applied for 20s, gently air-dried and cured for 20s) should then be applied prior to the addition of new resin composite.49 The presence of an adhesive was found to improve the strength of repair significantly (Figure 10).50

The mechanism of action of phosphoric acid is believed to be via removing surface contamination and increasing the reactivity of the silica within the composite, which promotes bonding with the silane coupling agent. As phosphoric acid etching does not result in significant increases in micromechanical retention, roughening with a bur is required prior to its application. Air abrasion using silica oxide particles results in mechanical roughening and a tribochemical silica coating promoting bonding to the silane.48^,49

With all repair techniques, wherever possible it is recommended to optimize bonding to enamel adjacent to defects.

Repairing silorane-based composites

In parallel with methacrylate-based resin composites, the different surface treatments for silorane repairs have been discussed in the literature.51 The most tested repair technique was the use of a silane coupled with an adhesive in common with the methacrylate-based resin composites.52 The protocol for silorane repair is similar to that of a methacrylate-based material and, as discussed earlier, it is beneficial, but not essential, to use the same type of material where possible.51 Baur and Ilie investigated the effect composites of different chemical composition (methacrylate-based, silorane-based and ormocer-based) when used to repair each other. Interestingly, the silorane-based composites showed the highest bond strength when applied to the composites consisting of different chemical composition.53

Resin-modified glass ionomer cements (RMGICs)

In contrast to amalgam and resin composite restoration repair, techniques to repair RMGICs are scarce on account of their use as a provisional restorative material. The bond strength and thus repair potential of RMGIC with RMGIC decreases with time. Direct repair with RMGICs, even with prior roughening with a coarse diamond bur and application of phosphoric acid or polyacrylic acid, showed marked variation between brands, indicating unpredictable results with RMGIC-RMGIC bonding.54

Resin composite bonding produces more predictable bond strengths and is considered a more reliable method of repair. Thus, conversion of the original restoration to a RMGIC–resin composite laminate/layered/'sandwich’ restoration is the method of choice. Modifying the surface of the RMGIC prior to application of the resin composite with a bur was superior to phosphoric acid etching.55

Replacement of RMGIC

The total replacement of RMGICs is a more likely clinical occurrence than temporary repair due to it being an unsuitable long-term restorative material. A minimally invasive approach should consist of removing the restoration and conforming, where possible, to the cavity design, which is likely to involve an adhesive restoration such as resin composite (Figure 11).

Glass ionomer cements (GICs)

The highest bond strengths of GICs were on newly set surfaces, which decreased with time, so the long-term viability of GIC repair is unknown.56

If the repair of temporary/provisional GIC restorations is necessary the application of phosphoric acid or roughening the surface followed by etching produces the highest bond strength prior to further GIC placement.57 Again, conversion of the original restoration to a GIC–resin composite laminate/layered/'sandwich’ restoration is often the method of choice, thanks to the more reliable bonding procedures between old GIC and new resin composite.

Summary

Utilizing the principles of minimally invasive dentistry (‘5Rs’: reviewing, resealing, refurbishment, repair and where necessary, replacement) encourages the preservation of tooth structure and maintenance of endodontic health, so extending the clinical life of the tooth-restoration complex. Current clinical evidence indicates that the ‘5Rs’ checklist described in this paper should be contra-indicated before justifying the total replacement of a failing direct restoration.