What is the limit for composite restorations?

Full disclosure, I am a prosthodontist, which means I have been trained in, and continue to train others in, the dark art of tooth preparation. But I have a confession. I believe tooth preparation to be a fundamentally bad thing. Just as GV Black's ‘Extension for Prevention’ philosophy is now rightly viewed with abject horror from our enlightened modern perspective, I believe that wilfully removing one of nature's most robust materials will be viewed with utter incredulity by generations of dentists to come. Does this mean I don't ‘do crowns’? Of course I do, but only in very specific situations: treatment of cracked tooth syndrome; replacement of failed existing crowns; and, very occasionally, as abutments for fixed or removable partial dentures. In all other situations, I believe that there are much more sensible ways of achieving our goals, whether to improve shade, shape, tooth position, occlusal function or protection of the remaining tooth structure.

So how did I come to this heretical conclusion? When I graduated in 1996, composite resin restorations were still relatively niche, and I certainly would never have considered placing any Class II composite as an undergraduate, let alone one that could do the job of a crown. My road to Damascus started when I took a job working from a mobile dental surgery in the Australian Outback in 1999 (Figure 1). There was no lab support for hundreds of miles, so direct was the only option. After returning to the UK, I started my postgraduate studies at the Eastman Dental Institute just as they published a landmark paper on the use of extensive direct composite restorations in the treatment of extensive anterior wear, adopting the Dahl Approach (Figure 2).¹ A few years later, the same unit published on the concept of assessing remaining structural dentine when planning for indirect restorations (Figure 3), highlighting how little tooth structure remains after crown preparation.² The marrying of these two concepts lead to a fundamental change in my philosophy, from a subtractive to an additive approach to restoring teeth. It is pertinent at this point to consider the common reasons why crowns are prescribed (Table 1).

Note that only the last two entries in Table 1 have no viable conservative alternatives. This is relevant when we consider that the survival of commonly prescribed all-ceramic crowns has been cited to be as low as 48% over 10 years.⁴ The fundamental problem is that the teeth being crowned are already depleted of structural tooth tissue, so the further tissue removal weakens, rather than strengthens, the tooth and reduces its lifespan. One suggested possible, but controversial, reason for the prevalence of the prescribing of crowns is that it may be in part due to the increased renumeration for crowns compared to direct restorations under many dental payment arrangements.⁵

A conservative alternative?

Unlike crowns, direct composite restorations do not rely upon traditional principles of retention and resistance form, and do not require removal of undercuts for placement. On the contrary, undercuts are positively encouraged as they provide macro-retention in addition to micro-retention produced by the acid-etch technique,⁶ formation of a dentine hybrid layer⁷ and, in recent decades, chemical adhesion to both enamel and dentine using molecules such as MDP.⁸ The scope of large direct composite restorations has increased further with developments such as deep margin elevation (DME) (Figure 11) and the introduction of mechanically superior materials like short-fibre reinforced composite (SFRC).

Making the complex simple

Being faced with an extensively depleted tooth can be a daunting prospect for those planning a direct composite restoration. Dealing with one issue at a time, in the correct sequence, will greatly simplify any complex problem.

The recommended sequence

1.

Removal of aetiology (diet, OH, splint)

The causes of the problem should be ascertained in order to rectify them. This will include a diet history and advice, plaque scores and oral health index, possible referral for causes of tooth surface loss and active prevention, such as sports guards, hard maxillary splints and increased topical fluoride application.

2.

Planning (‘failing to prepare is preparing to fail,’ Benjamin Franklin)

Identify which cusps need protecting, including any that have cusp base incomplete fractures (Figure 12) or have a height to thickness ratio greater than one. Pre-existing or planned root canal treatment requires cuspal coverage on most premolars and molars, especially where a marginal ridge has been lost. Careful assessment of periodontal, prosthodontic or endodontic prognosis will highlight any potential issues before embarking on complex, time-consuming restorative treatment.

3.

Isolation with rubber dam

There are very few situations where placing a rubber dam will not make the procedure less stressful and more predictable. To master placing rubber dam, it needs to be well taught and routinely practised. One exception is if there is a need for DME,^9,10 when it is often practical to carry out this part of the procedure before rubber dam application.

4.

Assessment of restorability

Once isolated, any pre-existing restorative material should be carefully removed to assess structural cracks (Figure 13), caries extent and pulpal exposure.

5.

Surface preparation

Following removal of all infected dentine, the entire cavity and adjacent sound tooth tissue may be sandblasted with 50-micron aluminium oxide to create a clean, roughened bonding surface, and to remove any loose enamel prisms that may separate on later polymerization of the composite resin. A three-step etch/prime/bond adhesive system is preferred by the author in order to optimise the bond strength to the individual substrates (Figure 11f).

6.

Rebuild missing axial walls

To create anatomical proximal surfaces, the use of curved sectional matrices is recommended. These require a separation ring to secure them and temporarily open-up interdental space. To achieve this effectively, two opposing axial walls are required. If one of the axial walls is absent owing to caries or fracture, then it should be first re-built using circumferential matrices or free hand (Figure 11g).

7.

Remove excess as you go

It is easy to lose anatomical landmarks when building up large direct composite restorations, since any excess material from one step in the procedure can then dictate where the next increment of material is placed. The use of a sharp No 12 surgical blade is very effective in trimming away this excess without contaminating the bonding surface.

8.

Proximal contacts and marginal ridges

Unlike Class II amalgam restorations, composite cannot be packed to create tight proximal contacts. To achieve this, the use of curved sectional matrices is required (Figure 11h). Overly forceful use of proximal wedges should be avoided as this can distort the matrix. The compression rings effectively tip the teeth, opening up the space to create a positive contact point. However, if both mesial and distal proximal contacts are being restored on the same tooth, they should be done one at a time, otherwise the tipping effect is cancelled out (Figure 14). The proximal wall should be placed in a single increment and adapted to the cavity margin using suitable equipment (Figure 11i). (This is the author's personal preference since it reduces the risk of creating voids/folds in impossible-to-access areas, although it is noted that the increased configuration factor may concern some). Verify the integrity of the contact point, if floss is not resisted, the proximal wall should be removed and replaced, correcting for any errors in the original placement.

9.

Structural dentine replacement

At this stage, you will effectively have a Class I cavity. The dentine can be replaced with conventional resin composite; however, I prefer to use an SLFC for dentine replacement owing to its possible superior structural qualities. This material is designed to be placed in bulk, with up to 5.5 mm curing depths for the translucent version (to fill root canal orifices and pulp chambers) or 2 mm for the dentine shade (Figure 11j). Notwithstanding this, it is desirable to cure resins in small increments to minimize polymerization shrinkage forces on the teeth.¹¹

10.

Cusp replacement

If cusps are missing, due either to caries or fracture, or have been reduced to provide cuspal protection, then the next step is to rebuild the external form of the cusp up to the tip using restorative composite resin. The positioning of the cusp tip should be done with reference to standard anatomical form and any individual characteristics exhibited by surrounding teeth (Figure 11k).

11.

Occlusal table

It is desirable to build up aspects of the occlusal anatomy, individually creating convex forms with deep fissures separating them. This is not merely because it looks natural and pleasing to the eye, but because the smaller increments limit any undesirable polymerization shrinkage and the deep fissures prevent the individual increments pulling adjacent ones away from the cavity walls. Once the anatomical build-up is complete, fissure stain may be sparingly applied where a sufficiently deep fissure exists, to blend with the surrounding teeth. Once you are satisfied that no further increments are required, an oxygen barrier such as glycerine or KY Jelly can be applied and a final cure completed to ensure full polymerization of the surface layer of composite (Figure 11l), although this uncured layer will be removed by finishing and polishing.

12.

Occlusal check

Convex anatomy provides multiple potential contact areas and is quick and simple to adjust to accommodate the patient's intercuspal position and avoidance of undesirable contacts in lateral excursive movements (Figure 11m–o). In contrast, flat or concave occlusal anatomy will be far more labour-intensive to adjust, and result in deep interdigitation of cusps that have the potential to exert harmful lateral forces to the restoration and tooth. It is essential to ensure holding contacts in intercuspal position rather than leaving the restoration out of contact, since the teeth are likely to erupt to regain occlusal contact, but the clinician will not then have the opportunity to adjust the harmful eccentric contacts.

13.

Finishing and polishing

If you have followed an additive approach and removed excess material at each stage, then there should be minimal need for finishing. Many approaches to finishing and polishing exist, but the following is the author's preference. Any flash can be easily removed with a sharp No 12 scalpel blade. Medium grit abrasive discs can be used at slow speed (3000rpm) to contour the axial walls and soften the marginal ridges. For occlusal adjustment, a fine rugby-ball-shaped diamond used at slow speed (20,000rpm) is a very efficient tool. Polishing can be achieved very quickly and simply. First, use a pre-polish spiral polisher (Eve Diacomp Plus Twist) dry at 10,000rpm (Figure 15a), to remove any excess over the cavity margins. Follow with a slurry of medium grit pumice in glycerine in a prophylaxis cup at 3000rpm to lift any excess adhesive and produce a satin finish (Figure 15b). A final long-lasting high gloss polish is achieved using a spiral polisher (Eve) dry at 10000rpm (Figure 15c), moving the polisher constantly over the surface.

14.

Maintenance

All effort to remove the causative factors should be made so use of fluoride and occlusal splints where indicated should be maintained indefinitely as well as reinforcing dietary and oral hygiene advice. Attention should be given to the anterior guidance, since wear to the anterior teeth may increase the loading on posterior restorations over time. Small amounts of wear and chipping may be seen, and can either be polished or repaired easily. Ultimately, the composite resin may fail, but the underlying tooth will remain intact to be restored in the future, not something that can be said of teeth destined for crown preparations.

Direct, indirect or silicone key?

The author's preference is for direct composite resin restorations, placed freehand using an incremental build-up in anatomical sections. In theory, an indirect approach could give greater control over the contours, but requires the removal, or blocking out of undercuts. In contrast, a direct approach allows engagement of all internal undercuts, as well as allowing the restoration to be placed over the exterior aspect of the tooth, engaging a much greater area of enamel. A stamp, or injectable approach, using a clear silicone key can simplify placement, but cannot create anatomical, cleansable contact areas (Figure 16). In posterior occlusal wear cases with intact contact points, an index for stamp or injection can be used. Neither indirect nor silicone key direct methods have the ability to deal with complex situations, such as DME, and so would be likely to require an additional visit to complete.

Time and money

Everyone will work at a different pace, but the time it takes me to complete a cuspal-coverage composite is approximately 75% of the time it takes me to prepare a tooth for a crown, but with no fit visit, no lab bill, a single local anaesthetic, no problems with temporization and possible benefits for the pulp. At worst, you have a very effective long-term temporary restoration that will serve as an excellent structural core for a crown should the need arise.

Finding the limit of composite restorations

Composite restorations have filled many niches that I simply couldn't have imagined when I started my career in dentistry over 30 years ago. They are the answer to many restorative problems, but not quite all. There are two situations where I believe that composites would not be my first-choice material. First, when effective moisture control is just not possible, such as with extensive root caries, then an alternative material may be considered. The second is when the tooth has a structural crack not limited to a cusp base, but traversing the cavity floor. In this situation, I prefer the rigidity afforded by an indirect cuspal-coverage restoration, preferably in a cast alloy. Although the ultimate survival of the tooth is still not assured in such situations, I have concerns that composite resin may not be sufficiently rigid to prevent further propagation of such structural cracks, despite views to the contrary¹² leading to the possible sequelae of pulpal involvement, or vertical root fracture (Figure 17).

Conclusion

During my 32-year journey in dentistry, I have seen the role of direct composite resins expand from small intracoronal restorations to a material capable of replacing the vast majority of coronal tooth tissue, encroaching on the traditional area of dentistry provided by indirect restorations. Controversial as it may sound, perhaps with its ability to be ultra-conservative and aesthetic, the only limit of this dental restorative material could be our ability as clinicians to use it to its full potential.