There can be confusion for clinicians when trying to decide if direct composite is the adequate solution to treat patients, or if dong an indirect restoration will ensure the best chances of a predictable result.

Especially in higher risk posterior areas, when we think about direct versus indirect restorations we want to pay very close attention to what the literature suggests will be a predictable result. There are four guidelines that I will evaluate in order to determine if a direct restoration will yield the most successful results.

This is not to say that direct composite can't be done if these criteria aren't met. However, if these criteria were met there would be very little reason to not use a direct restoration.

  1. Gingival margins in enamel: When the gingival margins of the proximal box are in the enamel, we get less recurrent caries in direct resin restorations as opposed to when the proximal box is all dentin. One of the problems with direct resin is shrinkage; when we cure it the composite shrinks and risks micro leakage. However, if there is enamel to etch and bond to the risk of leakage is significantly reduced.
  2. Isthmus is less than half the intercuspal width: The smaller the composite, the more tooth structure remains to resist functional forces and flexure. When the isthmus is less than half the intercuspal width, it means the cusps retain a significant resistance to flexure, which reduces the fatigue placed on the bond to the restoration. The wider the composite gets, the weaker the cusps become, meaning cuspal flexure becomes an increasing problem, this flexure can result in bond failure and the introduction of leakage.
  3. Ability to isolate and visualize placement: If you can't isolate, bond and visualize your placement, there is no way to do a predictable direct composite restoration. When you can't visualize and isolate, you should opt for a more forgiving material, such as amalgam or consider an indirect restoration that doesn't require perfect isolation such as a cemented restoration.
  4. Occlusion supported by enamel: Another factor that can add to the predictability of direct composites is if the occlusal contacts are on the actual tooth structure itself. This minimizes the risk of overloading the composite, reducing the risk of fracture and wear.


Commenter's Profile Image Lina
August 1st, 2013
Thank you Frank for this guidelines. Simple and clear.
Commenter's Profile Image Richard Featherstone
August 1st, 2013
Thanks for the great review. I like the part about keeping contacts in enamel, since I forget to consider it. I have the opinion that when a margin is in dentin in a deep box, that a resin-ionomer liner such as Vitrebond preceded Gluma or similar product will greatly reduce the micro leakage. It's also my opinion that an cementing an indirect restoration requires perfect isolation, equal to composite bonding for predictable success. I believe it is achievable with current tools available, like rubbing Ultradent Viscostat below the margins, and ultimately, but seldom, packing with cord.
Commenter's Profile Image Gerald Benjamin
August 2nd, 2013
I have to take exception to some of Frank's comments on posterior direct restorations. I placed a few thousand Concept (heat and pressure treated resin but resin none the less) and the overwhelming majority of these restorations exist 20 years later. This means that resin is capable of withstanding occlusal forces over long periods of time. The potential problems of resin shrinkage has been greatly overstated because resins in the 1980s and 1990s had significant amounts of shrinkage. Since the early 2000s nanohybrids and current resins have shrinkage rates below 2%. Bulk placement of Filtek Supreme Ultra and bilateral transenamel illumination using two powerful lights produces the least mount of shrinkage so that when placed correctly, shrinkage is not a problem. The overwhelming majority (95%) of dentists do not use a rubber dam hence they have virtually no hope of obtaining perfect isolation which is essential to clinical success of resins. Since 1992 with the introduction of dentinal adhesive dentistry, I have placed thousands of direct posterior resins with a survival rate of 95+%. The overwhelming majority of practicing dentists have NEVER taken a clinical course on the correct placement of posterior resins which leads to a high failure rate. Direct posterior restorations have a fabulous success rate if: 1. rubber dam is utilized 2. a proper adhesive protocol is followed 3. all cavo-surface margins are beveled with the final step of placement (pre curing) having the dentist pull the resin over the bevel with a sable brush . This means that there never is a definitive margin to leak. 4. bilateral transenamel illumination using two powerful lights simultaneously from the buccal and lingual with a final cure from the occlusal (60 seconds total cure time) My success with resin has not been dependent of size of the restorations (sorry GV Black) or whether the restoration is in enamel. At this point in time and with 20 plus years of history, the limiting factor for the majority of resin restorations is the clinician.
Commenter's Profile Image CARLOS MAS BERMEJO
August 7th, 2013
Very clear. Thank,s frank. I love this protocols
Commenter's Profile Image Mary Oz
August 7th, 2013
I totally agree with Dr Benjamin. I am a recent university graduate and would have to say that it all comes down to the margins of the restoration and isolation: both dependent on the eyes/skills of the dentist +/- assistant. I have a lot to develop and get better but I remain a great user of the rubber dam and proper isolation. I have been yearning to go listen to Dr Bertolotti talk but too far and costly for now; other than his courses on bonding, any suggestions? Thanks!
Commenter's Profile Image Michael Ogden
August 17th, 2013
I agree with Dr. Benjamin with a couple of caveats. I see no need to bevel cavosurface margins. I do so at the gingival margin only to facilitate sectional matrix placement. On occlusal surfaces, I use a brush tip moistened with Ultradent Permaseal to blend the margin and reduce the oxygen-inhibited layer. One more tip--use a drop of flowable spread along the gingival margin and internal line angles and add a nanofilled composite without curing the flowable first. It will act as a wetting agent and doesn't become a substantial part of the restoration, but absolutely eliminates voids. You can use it the same way for subsequent increments. My observation is also that both microhybrids and nanohybrids resist fracture and occlusal wear to the point where we don't have to worry about that anymore. I can't remember the last time I saw a posterior composite fracture, honestly.
Commenter's Profile Image Gerald Benjamin
December 21st, 2013
Hi Dr. Odgen; I have learned one very important lesson during my almost 40 year career: Follow the protocols of the best in the world. Having taken many courses with Newton Fahl, the undisputed best resin dentist in the world, he strongly recommended beveling all margins 12 years ago. The first 10 years placing direct posterior resins, I did not place bevels on the cavosurface margin but still pulling the resin over the margin with a brush. For the last 12 years, I have placed a cavosurface margin bevel and the restorations with the bevel look better than the non beveled restorations. Granted, they all look good with zero leakage. There may be some marginal wear but NEVER leakage. Anything done for more than 10 years on a daily basis in dentistry is beyond anecdotal. (Aside: When I began my studies with Frank almost 20 years ago, I did not listen to his wise advice to never be in the first group to use a new material…I learned the hard way to listen to the best. That is WHY they are the best.)
Commenter's Profile Image Hisham Osama
March 13th, 2015
Terrific analysis for wide range of cases based on sound and clear scientific facts which ensures fast and sharp decision making
Commenter's Profile Image Hisham Osama
March 13th, 2015
The article reflects wide,free and supreme dental knowledge with connecting various aspects "cavity design , bonding , occlusion" all together in simple,short and heavy weighed words. Go on for more success