Ceramic onlays: indications and adhesive principles

14 February, 2022 / clinical
 John McQueen  


In modern-day dentistry, the clinician is afforded a vast armamentarium of restorative options for a carious, failing, or fractured tooth. Advances in adhesive technology and a move towards more conservative options have made the onlay an increasingly popular treatment modality. In the following case study, we will discuss the indications and benefits of the ceramic onlay vs other treatment options, as well as demonstrate the gold standard protocol for preparation and cementation.


Direct restorations should be limited to smaller cavities, generally, those that are one-third or less of the tooth’s bucco-lingual width. If a cavity isthmus extends beyond this, a large degree of functional demand is placed on the restoration1. In these scenarios, the large isthmus generally correlates to thin, weakened, cusps. To avoid fracture, an indirect restoration should be considered. A classic clinical example of this would be an old amalgam restoration with a large isthmus +/- a cracked/fractured cusp. 

In the past, the go-to option for a tooth of this description would be a full-coverage crown. However, dentists must aim for the most minimally invasive option available. In 2002, Edelhoff and Sorenson published a study showing that a classical preparation for an all-ceramic crown (1.5mm occlusal reduction and 1mm margin) on an upper first molar resulted in the removal of 72.3% of tooth structure.

In comparison, an MOD onlay preparation (1.5mm functional cusp, 1mm non-function cusp reduction) only removed 39%2 – almost half as little reduction as the crown. Research carried out by Frankenberger et al in 20083 and Van Dijken et al in 20104 showed ceramic onlay survival rates to be 84% at twelve years, and 76% at fifteen years, respectively. With regards to the incidence of pulp death, dentists classically quote 15-20% risk with full coverage crowns5, 6. Whereas studies have shown, over fifteen years, the risk to be as low as 0.8% with ceramic onlays.4

Considering the evidence base, a ceramic onlay provides dentists with an indirect restorative option that can protect weakened teeth for a length of time similar to that of a full-coverage crown. Whilst doing so, its preparation can be twice as conservative, with risks to pulp heath a fraction of those offered by crowns.

Case study

A fit and well 60-year-old man presented for a routine examination, and the following findings were noted in the upper right quadrant [Figure 1].

Figure 1: The patient had two large class II amalgam restorations present in teeth 16 and 17.

The patient had two large class II amalgam restorations present in teeth 16 and 17. On further inspection, there were ditched margins, signs of fracture lines running from the restorations, secondary caries noted around the MB cusp of 17, and there was a significant overhang on the amalgam of 17. After careful deliberation, we opted to remove these failing restorations along with any caries and planned to restore both teeth indirectly with Emax onlays.

Rubber dam was placed before removing any old dentistry. This would ensure a clear and dry operative field throughout the procedure. Once the failing restorations and any infected dentine were removed, the cavity was cleaned with the CoJet air particle abrasion system [Figure 2].

Figure 2: The cavity was cleaned with the CoJet air particle abrasion system.

This will increase the bond strength to both healthy tooth tissue and affected dentine7, 8. Now thoroughly cleaned, the remaining tooth structure could be assessed to allow planning of the restoration’s parameters. 

It was decided at this stage that Deep Marginal Elevation (DME) would be carried out on the distal margin of 16 and the mesial margin of 17 respectively [Figure 3].

Figure 3: Deep Marginal Elevation (DME) was carried out on the distal margin of 16 and mesial margin of 17.

DME is useful, not only as it gives the operator full control over the shape and position of the definitive margin of the prep, but it also increases the ease of rubber dam placement for the fit appointment by raising the margin 2-3mm supra-gingivally. In this case, each margin was treated individually by placing a 50 micron-thick sectional matrix and an anatomical wedge. The enamel margin was etched for 20 seconds using 38% ortho-phosphoric acid, thoroughly washed, and then a two-stage self-etching adhesive (Clearfil SE) was applied to the box. The margins were raised, and the boxes flattened out using incremental layers of a heated nano-hybrid universal composite (Venus Diamond)9.

Now that the cavity margins are all supra-gingival and the proximal boxes are flattened out, the preparations can be completed. An Iwanson gauge was used to assess cusp thickness and any cusp <2mm thick was reduced by 2mm and incorporated into the preparation. Care is taken to remove any undercuts and to round out and smooth any sharp internal line angles. The enamel margins are bevelled to a 45° angle using a medium-grit rugby ball diamond – this exposes a greater surface area of enamel rods, therefore increasing enamel bond strength10, and aids in marginal transition for the final restoration’s aesthetics.

The final, and arguably most important, stage before impression making, is Immediate Dentine Sealing (IDS)11. Applying a dentine bonding agent (DBA) to freshly prepared dentine ensures the surface is free from contamination, therefore creating the ideal circumstances for dentine bonding12. IDS allows pre-polymerisation of the DBA, leading to an increase in bond strength to dentine13. In the case of indirect restorations, the bond matures and strengthens in a stress-free environment beneath the provisional restoration14.

In 2007, Pascal Magne’s article in the Journal of Prosthetic Dentistry showed the microtensile bond strength of Clearfil SE bond to be 55.14MPa when IDS was carried out with a two-week delay in final restoration cementation. This is over thirty times stronger than if the dentine bond was delayed to the day of the fit (1.81MPa)15. In this case, the dentine and DME were cleaned once more with air particle abrasion and the two-stage adhesive of Clearfil SE bond was applied and polymerised [Figure 4].

Figure 4: The dentine and DME were cleaned again with air particle abrasion and the two-stage adhesive of Clearfil SE bond was applied and polymerised.

A second cure was completed with a glycerin barrier- this ensures polymerisation of the oxygen-inhibited layer and will also aid in the removal of the provisional at the next visit15.

With the preparation now complete [Figure 5], the dam is removed and definitive on impressions were made and sent to the laboratory for construction of two Emax onlays. Cavit was used as the first layer of provisional restoration (packed into the residual boxes to aid ease of removal) and then the preps were shrink-wrapped a bis-acryl provisional material (Quicktemp 2).

Figure 5: With the preparation now complete, the dam was removed and definitive on impressions made and sent to the lab for construction of two Emax onlays.

Two weeks pass, allowing our dentinal bond strength to mature and our lab to construct the definitive restorations. Like the preparatory stage, the fitting visit was completed under rubber dam to provide gold standard isolation for bonding. After the fit-check is approved, the intaglio surfaces of the restorations are etched with 9% hydrofluoric acid for 20 seconds, silanated with a two-step silane coupling agent (bis-silane), and then bonded with an unfilled, HEMA-free, hydrophobic resin (heliobond)16.

This is an essential stage in aiding adhesion to silanated ceramics. HEMA is designed as a wetting agent which will draw water into the adhesive – ideal for bonding to moist dentine but will weaken the bond when bonding to a hydrophobic substrate, such as Emax16. Once the provisionals were removed, the preparations were, once again, cleaned with CoJet air particle abrasion. As well as removing any residual temp material this “reactivates” the IDS13.

The enamel is etched as it was for the DME stage and then the entire prep is coated with the second bottle of Clearfil SE and air-dried. It is important not to cure the bond at this stage because if any shallow pools of resin polymerise, they may stop the restoration from fully seating! In this scenario, the cement of choice was a dual-cure composite (Variolink neutral DC). A full chemical cure was allowed to complete, making excess removal much more straightforward. Once all excess was cleared the restorations were cured for 20 seconds on each fitting surface. As with the IDS, a final polymerisation was completed with a glycerin barrier15. [Figure 6].

Figure 6: Once all excess was cleared the restorations were cured for 20 seconds on each fitting surface and a final polymerisation was completed with a glycerin barrier.

In conclusion, advances in adhesive and biomimetic dentistry have afforded clinicians excellent predictability in bond strength whilst maintaining tooth structure with conservative preparations. This type of dentistry requires a sound understanding of the protocols, a selection of high-quality materials and dentine bonding agents, and, above all else, the dexterity and patience to follow the gold standard. When executed correctly it can have a highly successful and deeply satisfying result for both the clinician and the patient.

John McQueen is a restorative dentist based in Advanced Dentistry @ Hyndland Dental Clinic and Lockerbie Dental. He has a keen interest in contemporary, adhesive dentistry with a focus on minimally invasive methods.


1Jackson, R. Esthetic inlays & onlays: Contemporary Esthetic Dentistry. Ed. Freedman, G. Chicago: Elsevier, 2012: 469-481.

2Edelhoff, D. & Sorensen J.A. 2002. Tooth structure removal associated with various preparation designs for posterior teeth: Int J Perio Rest Dent; 22: 241-249.

3Frankenberger R, Taschner M, Garcia-Godoy F, Petschelt A, Kramer N. 2008.

Leucite-reinforced glass ceramic inlays and onlays after 12 years. J Adhes

Dent. 10(5): 393–398.

4Van Dijken, J.W.V.; Hasselrot, L. A Prospective 15-Year Evaluation of Extensive Dentin–Enamel-Bonded Pressed Ceramic Coverages. 2010. Dent. Mater; 26: 929–939.

5Cheung, G. S. P. Fate of vital pulps beneath a metal-ceramic crown or a bridge retainer. 2005. Int J Endo; 38: 521-530.

6Saunders, W.P & Saunders, E.W. Prevalence of peri-radicular periodontitis associated with crowned teeth in an adult Scottish subpopulation. 1998. BDJ; 185: 137-140.

7Van Merbeek, B, DeMunck, J, Mattar, D, Van Landuyt, K & Lambrechts, P. Microtensile bond strengths of an etch and rinse and self-etch adhesive to enamel and dentin as a function of surface treatment. 2003. Oper. Dent; 28 (5): 647-666.

8Sattabanasuk, V, Burrow, M.F, Shimada, Y, Tigami, J. Resin adhesion to caries-affected dentine after different removal methods. 2006. Aust Dent J; 51 (2): 162-169.

9Magne, P. Deep margin elevation: a paradigm shift. 2012. Am J Esthet Dent; (2): 86-96.

10Opdam N, Roeters JJ. Necessity of bevels for box only class II composite restorations. 1198. J Prosthet Dent; 80 (3): 274-279.

11Jayasooriya, P.R, Pereira, P.N, Nikaido, T, Tagami, J. Efficacy of a resin coating on bond strengths of resin cement to dentin. 2003. J Esthet Restor Dent; 15: 105-113.

12Paul, S.J, Scharer, P. Effect of provisional cements on the bond strength of various adhesive bonding systems on dentine. 1997. J Oral Rehabil; 24: 8-14.

13Ozturk, N, Aykent, F. Dentin bond strengths of two ceramic inlays after cementation with three different techniques and one bonding system. 2003. J Prosthet Dent; 89: 275-281.

14Dietschi, D, Monasevic, M, Krejci I, Davidson, C. Marginal and internal adaption of class II restorations after immediate or delayed composite placement. 2002. J Dent; 30: 259-269.

15Magne, P & Cascione, D. Immediate dentin sealing supports delayed restoration placement. 2007. J Prosthet Dent; 98: 166-172.

16Betancourt, D.E. Resin-Dentin Bonding Interface: Mechanisms of Degradation and Strategies for Stabilization of the Hybrid Layer. 2019. Int J Biomaterials; 2: 12-24.

Categories: Magazine

Comments are closed here.