The Injection Molded Class II Restoration

Dr. David J. Clark


Figure 1. Illustration of the Clark Class II restoration (left), slot prep (middle), and GV Black-style prep (right).

Stop Layering, Start Injection Molding
I first wrote about injection molding in 2010, lamenting that Class II composites were in the middle of a fitful transformation. Eight years later, we are still in the midst of chaos. Why? There is lack of consensus in dental schools, an upheaval in filling materials, and incessant marketing campaigns from manufacturers claiming all sorts of things. It has led many dentists to a state of discomfort or even paralysis, resulting in resistance to change.

It is not unusual to see significant disagreement on every aspect of the Class II restoration within a school, as faculty members undermine one another when talking to students. “I’ll wait until the dust settles,” is what I am commonly told. At the core of the problem is a cavity preparation design that originated in 1890. We are essentially handcuffed to the past. However, 2 dental schools are now teaching modern, non-retentive, infinity edge, Clark Class II cavity preparations and true injection molding: the Roseman University of Health Sciences College of Dental Medicine (Roseman) in South Jordan, Utah, and the University of Alberta (via a GPR program). In total, 6 dental schools plus government programs are sending faculty members to the Educator’s Summit this June at the Bioclear Learning Center in Tacoma, Wash, to train faculty on the Bioclear Method and collaborate on the research and logistics needed to work with licensure examination boards to eliminate the tragically outdated GV Black cavity preparation when placing Class II resin-based restorations for clinical competency exams. A few other schools are considering eliminating layering and boxy preps, as they have dismal results compared to the monolithic, injected (bulk fill) method and there are a large number of fracturing teeth in the population.

Other important steps foward include studies demonstrating the superiority of these modern techniques and preparations. A study is underway at the Minnesota Dental Research Center for Biomaterials and Biomechanics that will compare traditional cavity preparations to the new Clark Class II. The Dalhousie University Faculty of Dentistry and The University of North Carolina at Chapel Hill School of Dentistry are also performing studies on different aspects of the method. In addition, a clinical study comparing layered GV Black preps against non-retentive Clark Class II Preparations (Figure 1) is in progress at Roseman. This study compares the old and new, with the exiting fourth-year students cutting boxy preps and cold layering as third-year students cut the nonretentive Clark Class II restorations using the injection molding technique with a combination of heated Filtek Bulk Fill Flowable Restorative (3M) and heated Filtek One Bulk Fill (3M) (paste) composite. This study is a 3-way collaboration between the school, the Bioclear Learning Center, and Technologies in Restorative and Caries (TRAC) Research.

The New Lexicon of Class I and Class II Restorations
Moving forward, it is critical that we all speak the same language. To that end, we are committed to the use of a modern and consistent lexicon. The first and probably most important new name (or descriptor) is that of the Infinity Edge Tooth Restoration Interface (TRI) (Table 1). The pioneers of composite resin dentistry junked the concept of a definitive margin and began teaching an infinity edge in anterior teeth decades ago. Today, we update that concept again with the TRI and extend it, where possible, beyond the bevels and flared walls onto blasted, etched, and uncut enamel. It is as desirable in posterior restorations as it is in anterior restorations where our bias tells us it’s “okay.” The TRI is a zone of interface that is a luxury afforded for enamel adhesion. In deeper restorations with the TRI on cementum, it becomes more like a traditional “margin” with a chamfer shape preparation. Some apical creep of the material onto clean-blasted cementum, to a very limited extent, is acceptable and, frankly, inevitable. Essentially, we plan to do what the composite wants to do and will do anyway: to go on and around the tooth rather than in the tooth, with composite extending beyond the futile containment at an actual frank margin. The modernization of the old “margin,” ie, the TRI, changes the game.

In Enamel We Trust
Today, we plan for and encourage the injected composite to flow unencumbered to surround the tooth (Figures 2 to 13). These areas (if properly prepared by blasting and sanding, etching, adhesion, and injection molding), with significant engagement of enamel rods opened at 45° to 90° to the enamel rod orientation, are essentially bulletproof. This combination of additive dentistry (composite beyond the cavity area extending to the line angles), saucer preparation with double serpentine radius bevels and avoidance of any mechanical retention into dentin, and a compression joint on the occlusal primary strike point via an aggressive radius bevel changes the game significantly. The low regional C-factor approaches a ratio of one (or less) as the composite stops fighting itself during polymerization and acts more like a porcelain laminate mostly bonded to enamel.

Repeated studies have shown that porcelain veneers bonded to enamel can have a 95% survival rate at 10 to 15 years.1,2 Veneers bonded to half enamel and half dentin were less successful. Despite advances in dentin bond strengths, the long-term performance of dentin bonding is significantly inferior to enamel bonding when the restoration is placed in tension. The big picture is that we need dentin bonding for desensitization and to keep those areas from leaking, but only enamel adhesion can be trusted to keep a restoration from failing when loaded in tension. Because of the endemic overreliance on dentin adhesion, schools chase their tails as the restorations involving slot preps fall out at about year 5, and then they often go back to being “tooth carpenters,” using GV Black preps with occlusal dovetails and dramatically weakening the tooth while providing good mechanical resistance to dislodging. Is this the best we can do?

Figures 2 to 13. This clinical case demonstrates a quadrant of incipient Class II restorations restored with nonretentive saucer preparations on the interproximal and fissurotomy preparations on the occlusal. (The disclosing and blasting of biofilm, which was performed after the placement of a rubber dam, is not shown here.) “Pre-wedging” is described as placing the Diamond Wedges (Bioclear) before cutting the preparations. The final step of cavity preparations is sanding the interproximals with a lightning strip (Integra Miltex) and then a final round of blasting. The quadrant is matrixed with Biofit HD molar and premolar matrices, re-wedged with the same Diamond Wedges, and then the TwinRing separators (Bioclear) are placed as we start the injection molding of the middle tooth. For a complete summary of the case and a more complete description of the injection molding technique, please view the Biofit video at The completed case shows the infinity edge ensconcement of the teeth with leak-resistant, stain-resistant, and fracture-resistant form. Note that the composite is allowed to flow to the line angles of the tooth for the modern “extension for prevention.”

Monolithic Composites: The Next Big Thing
In Figures 14 to 16, I have colored 2 of the 3 resin components: The adhesive is colored orange and the flowable is colored green. The regular composite is a normal color. I then layered the first dentoform tooth with a popular technique: using an initial layer of flowable that is cured, followed by a second “capping” layer of regular (paste) composite. The final results (Figures 17 and 18) demonstrate several key differences between the old and new methods. Monolithic injection molding dramatically reduces voids and seams. Heating the composite (Figure 19) reduces viscosity and voids, increasing strength as an additional benefit. As with all-ceramics, monolithic composites are stronger.

Do We Ever Need to Layer?
At the Bioclear Learning Centers, we insist that all posterior restorations be restored with a bulk fill flowable composite plus a bulk fill regular composite, managing them much like a light and heavy body impression technique. Filtek One Bulk Fill is as strong and aesthetic as the Filtek Supreme Ultra Universal Restorative (3M) that we use for anterior restorations, so there is no reason not to use a bulk fill. If the occlusal area is deeper than 4.0 mm, or the interproximal is deeper than 5.0 mm, then we layer for depth-of-cure reasons by putting down and curing a gingival increment of bulk fill flowable so that the remaining space can be subsequently injection molded and cured. Otherwise, in general, we do not layer.

Translucent Matrix System
Injection-molded composite dentistry benefits from a new kind of matrix system that includes the use of translucent matrices (Biofit [Bioclear]). Traditional metal matrices allow curing only from the occlusal surface. Clear matrices make it possible to perform buccal lingual curing with the matrix in place and also facilitate a much healthier contact because metal can only be shaped so much.

Figures 14 to 18. This series of images compares layering with a metal matrix and injection molding with an anatomic mylar matrix (Biofit [Bioclear]). Remember that with injection molding, the tooth is etched, and then a universal adhesive (such as Scotchbond Universal Adhesive [3M]) is massaged into the dentin tubules, air thinned, and light cured. Injection molding comes next. First, a small amount of adhesive is splashed into the cavity and air thinned, then immediately followed with injected and heated Filtek Bulk Fill Flowable Restorative (3M) and immediately followed with injected and heated regular bulk fill composite (Filtek One Bulk Fill [3M]). Next, all 3 are cured as one monolithic mass, taking advantage of 3-point curing through the clear matrix in the interproximal. Layering is problematic, and studies show layering results in voids a staggering 70% of the time. When performing injection molding, the lesser resins, both adhesive and flowable, mostly act as temporary wetting agents and, with proper injection technique, the heavier resin displaces the lesser ones and they are wiped away from the occlusal before light curing. The goal should be to create a restoration in which 90% or more of the volume of the restoration is regular composite and only 10% or less is the flowable resin. In Figure 16, the dentoform tooth is pictured after placing an excessive amount (for demonstration purposes only) of the bonding resin that has been dyed orange. The canula of the flowable composite is poised to be injected into the pool of bonding resin. The green flowable is shown next mixing with the pool of adhesive. Figures 17 and 18 demonstrate that with a metal matrix, the surface is less smooth than what Mylar can impart. In addition, the obvious merits of not layering are shown.

Biofilm Identification and Blasting
Because we are allowing the composite to go beyond the terminus of the radius bevels, the enamel past the bevels has not been cut and will thus be covered with tenacious biofilm. Traditional scaling and prophylaxis with pumice and a cup rarely removes all biofilm. Also, phosphoric acid does not remove biofilm. It is no wonder that so many margins stain. One of the foundational components of the modern Class II is disclosing and then blasting the biofilm away with pressurized air-water-mild abrasive slurry. The Bioclear Blaster or a similar unit is key.

In this brief article, I have outlined the recommendations that have been provided to dental schools and to practicing dentists who have attended courses at Bioclear Learning Centers that are teaching certification classes in the Bioclear Method.


  1. Friedman MJ. A 15-year review of porcelain veneer failure—a clinician’s observations. Compend Contin Educ Dent. 1998;19:625-638.
  2. Ge C, Green CC, Sederstrom DA, et al. Effect of tooth substrate and porcelain thickness on porcelain veneer failure loads in vitro. J Prosthet Dent. 2017 Dec 19. [Epub ahead of print]

Dr. Clark maintains a private practice in Tacoma, Wash, and is the founder of the Academy of Microscope Enhanced Dentistry. He is also a course director at the Newport Coast Oral Facial Institute in Newport Beach, Calif, and he and Dr. Jihyon Kim are co-directors of the Bioclear Learning Center in Tacoma. Dr. Clark’s main areas of interest include the redesigning of restorative preparations and endodontic access preparations. He can be reached at

Disclosure: Dr. Clark is the owner of Bioclear Matrix Systems.

Related Articles

Composite Vs. Porcelain: What You Need to Know

Incommensurability in Endodontics: The Role of the Endodontic Triad

Management of Deep Caries With Bioclear and the Modified Hall Technique