: C C C
. .
+7 (383) 304 90 04
, , 12/3
9:00 18:00, -

Introducing the Clark Class I and II Restoration

Author: David Clark, DDS

Twenty-five years after their inception, posterior composites remain unpredictable. In comparison to amalgam restorations, posterior composites show significantly higher failure rates, are more costly, take longer to place, have more postoperative symptoms, leak, stain, chip and cause food impaction (Figs. 1-3). My former operative dentistry instructor (a legend in the Pacific Northwest) continues to build his legacy today at the University of Washington School of Dentistry. He recently shared this opinion with mehe hates posterior composites and hates to teach the technique. In this article, I will explore a technique that addresses the most serious shortcomings of posterior composite restorations.

Whats Wrong with this Picture?

Between 2005 and 2007, I lectured alongside Dr. Gordon Christensen and Derek Hein at the CRA Dentistry Update. A good cross section of restorative dentists from around the USA and Canada attend these lectures. At each city (and also at my private lectures) I have asked attending dentists (now numbering more than 6,000 clinicians) this question: How many of you feel that posterior composites are holding up as well as amalgams? Only a few hands go up. Over 95% have concluded that posterior composites are inferior to amalgams, yet most have stopped doing amalgam restorations and are placing posterior composites. Then I ask the attendees, So if amalgams are better, why are we doing posterior composites? The answers generally involve hand wringing, frustration, and lively debate.

Endodontists joke that posterior composites are the number one killer of pulps, that leaking composites are their number one referral source. Most studies have shown that Class I and Class II composites have a significantly higher failure rate than amalgam restorations. 1-4 The AMA, ADA, FDA, US Public Health service, CDC, NIH, and WHO have all declared amalgam safe. In light of this evidence and overwhelming opinion, how can we in good conscience continue to place posterior composites? So lets ask one more time, whats wrong with this picture? Why are we content to provide a posterior composite restoration that essentially cripples the tooth in the name of esthetics, knowing that there is no proven systemic health benefit compared to amalgam? Many restorative dentists have simply given up on amalgam and composite and spend more than $100,000 for a Cad Cam unit. They choose porcelain inlays and onlays as successors to amalgam rather than struggle and compromise with posterior composites.

The current state of posterior composite restoration poses empirical arguments for amalgam or porcelain alternatives. But by the conclusion of this article I will provide two very good reasons why posterior composites can bless, rather than curse the tooth.

The G. V. Black Era

To understand how clinicians can be influenced by a cultural and scientific icon, it is helpful to look at medical history. Claudius Galen was a Roman physician who boldly devised a medical model that doctors followed for fourteen centuries. Though his medical judgments were remarkably advanced for his day, today, not surprisingly, we know that most of Galens theories and treatments were completely wrong, and that the rest were seriously flawed. In a parallel to Galens example, G. V. Black was a consummate dentist/scientist and his exquisite designs for cavity preparation were a huge step forward for dentistry. Unfortunately, we are discovering today that those cavity shapes weaken the posterior dentition and lead to fracturing in even the most conservative applications5,6 (Fig. 4). 


Figure 1Whats wrong with this pic- ture? These recently placed posterior composites demonstrate the often woe- ful state of direct composite restorative dentistry. Black arrow: A carious fissure was missed; lack of magnification was the likely culprit. Blue arrow: Minimally invasive Class II cavity shape creates impossible C Factor problems. Red ar- row: Incremental loading leaves seams and voids that allow subsequent frac- ture. Green arrow: Proximal tooth was iatrogenically gouged and now has a carious lesion penetrating into dentin.

In a two-year study planned for future publishing; between 2001- 2003 I utilized 16X magnification to evaluate each posterior tooth that was treated for replacement of an amalgam or posterior com- posite. I classified and documented incomplete fractures. I found:

  • Sharp internal line angles are only a small part of the problem  
  • Joining the occlusal to the in- terproximal is the worst pos- sible design for crack avoid- ance and the most common area for crack initiation
  • Most fractures initiate in dentin at the line angles 
  • Interrupted cavities were more crack resistant than connected cavity preparations 


Figures 2 & 3are higher magnification views of Figure 1. 

In 2004, I collaborated with Dr Sheets and Paquette to publish the first guide to early and treat- ment of microscopic dentin and enamel cracks/fractures in the Journalof Estheticand Restorative Dentistry. In the May 2007 edition of Dentistry Today, (The epidemic of cracked and fracturing teeth), I highlighted the widespread problem of amalgam tooth prepa- rations that predispose the tooth to fracture. One of dentistrys myths is that amalgam expansion causes tooth fracture. Expansion failures have never been proven. The fracture problem does not originate with amalgam, per se. It originates in iatrogenic GV Black cavity preparations. And just as many of us feared, we are see- ing the same pattern of fractur- ing in teeth with posterior com- posites now that enough time has elapsed to assess their longevity.

The Simonson Transition

Dr. Richard Simonson is widely recognized as a pioneer in new cavity preparation shapes for minimally invasive, bonded, resin- based posterior composites. In spite of his innovations, the GV Black preparations that I was taught in school twenty years ago have been only slightly modified for posterior composites in the typical dental practice and in most dental schools.  4.png    

Figure 4Extracted bicuspid with a con- servative G. V. Black Class II features classic iatrogenic fracturing pattern.


Figure 5The gingival margin in this class II composite demonstrates the pervasive problem of microleakage. There was unfavorable C factor at the margin creating suck back. Uncured and contaminated flashing results from a metal matrix that blocks light curing and visualization.


Figure 6Case in Figure 5 is retreated with Clark Class II filling techniques, the clear and anatomic Bioclear matrix, combined flowable and paste micro- filled composite, to overcome multiple problems.


Figure 7A Bioclear molar matrix dem- onstrates the rounded, biomimetic shape necessary for ideal contacts and an ideal, infinity edge margin.


Figure 8Diagrams of the Clark Class II (5a), the slot preparation created by Simonson and others (5b), and the origi- nal G. V. Black class II (7c).


Figure 94X and 24X magnification comparing the tip size of the Fissurotomy NTF Bur versus a #556 fissure carbide. The tip size of the Fissurotomy NTF is easily ten time smaller. The width and length of the cut into dentin appear to be the most debilitating parameters predisposing posterior teeth to fracture. Sharp line angles (as produced by the #556 carbide bur) are not recom- mended for composite restorations. To allow for proper dispersion of the forces of mastications, all internal lines angles should be slightly rounded. Rounded internal line angles are created with the use of the Fissurotomy Burs. The Fissurotomy NTF Bur has the narrow- est tip available to completion of these procedures.

Table 1 Problems associated with current posterior composite    

Placement techniques 

Table 1 highlights problems with traditional posterior compos- ite techniques.

 A flat metal matrix, traditional wedge, boxy cavity shapes, biofilm that is difficult to remove just past the margins and incremental composite loading combine to give the common result in figure 5. The Clark Class II shape, aggressive sanding of the interproximal with a lightning strip prior to placing the matrix, an anatomic, translucent matrix (Bioclear Matrix System) that allows the composite to form an ideal feather edge, and injection molded single phase composite placement combine for a superior result (Figs. 6 & 7).  Composite is a poor biological space filler. A biological space filler such as amalgam or gold foil does not require any adhesion to the tooth surface. Composite on the other hand must be sealed 360 degrees and from inside to out.  Unlike amalgam and gold foil techniques, packing composite into a hole is not a predictable method. Excellent clinicians have been dealt an unfair hand when it comes to Class II composites. Most of the features of the traditional cavity preparations such as parallel walls, resistance and retention form work against posterior composites. What we have observed at CRA and under the microscope is that polymerization shrinkage cannot be eliminated, only mitigated. The best margin is no margin, and when composite extends slightly past the cavo-surface margin, it is generally well sealed with no white line. When we polish back to the margin, the white line often appears. Composite sealing with thin resins applied after filling the cavity may reduce wear. However, trying to seal an imperfect margin after the fact is futile. As I have explored these white lines, they generally extend completely to the pulpal floor, far beyond the reach of a sealer.


Figures 10 A-DA: Sectioned molar 4X and 24X magnification reveals a serious enamel defect that extends very near the dentin. This insidious defect is a perfect example of the unpredictable nature of occlusal morphology. The deep groove full of bioflim and caries activity is not part of the central groove and is also at an oblique angle to the long axis of the tooth. B-C: Initial penetration is achieved with the Original Fissurotomy Bur. The more aggressive taper on the 2.5mm cutting surface of this bur allows a conical access to the enamel defect. This shape affords proper visualization of the myriad of directions that are pos- sible. The most insidious types are the lateral and cul-de-sac type of defects that are often not discovered when parallel-sided cuts and made, compounded by insufficient magnification. D: Once the extent of the defect and or lesion is ascertained, the more delicate and less tapered Fissurotomy NTF Bur can be utilized to finish removing stain and bacteria. 


Figure 11The Cala Lilly is a graceful flower with an ideal shape to mimic for Class I cavity preparations. It allows tooth splinting when composite is care- fully placed; it maximized enamel rod engagement, allows perfect visualiza- tion of the cavity, and minimizes the potential for crack initiation.   


Figure 12Preoperative view of an ev- ery day occurrence. A decade after amalgam was placed; the tooth is dete- riorating as the fracture spreads.


Figure 13Composite placed in a tra- ditional parallel walled cavity cannot splint the coronal tooth structure. In contrast, when the Calla Lilly shape is created, we will engage nearly all of the enamel rods of the occlusal surface. Note that the caries had followed the crack into the mesial interproximal. C factor has been oversimplified and remains a significant problem. 

Table 2 The disconnected occlusal portion of the restoration can be:

Small defects; Fissurotomy shaped;
Moderate lesion; Cala Lilly shape
Large sized lesion or amalgam replacement; cusp tip to cusp tip
Splinting Cala Lilly shape 

Posterior composites should go on not in the tooth.

Minimally Traumatic dentistry should be considered as an upgrade of Minimally Invasive dentistry. Well meaning dentists are promoting minimally invasive dentistry. The best long term outcomes are more important than the race to minimize the micrograms of tooth structure that are removed. For example, the tunnel preparation preserves the enamel of the marginal ridge but unnecessarily weakens the tooth and impedes clinical visualization. Incomplete caries removal combined with excessive tooth weakening are unacceptable casualties of the noble mission to save marginal ridge enamel. 


Figure 14Clark Class II is claustro- phobic without magnification and unre- storable with conventional tools. With advanced magnification and new tools, we have new possibilities. Bioclear Average Curved Molar-ACM ma- trix pictured. Also shown is Original Interproximator in lieu of a wedge. 


Figure 15Low magnification view of postoperative result.


Figures 16 & 17High magnification post-operative views.


Figures 16 & 17High magnification post-operative views. 


Figures 18 & 19Low and high magnifi- cation of one year follow up.


Figures 18 & 19Low and high magnifi- cation of one year follow up.


Figure 20SEM of three year follow up of one my cases evaluated in the CRA study. Filtek Supreme paste combined with flowable technique in a molar shows good wear resistance. White arrow shows paste composite/flowable composite interface. Red arrow shows flowable composite/enamel interface. 

The Fissurotomy and Cala Lilly Class I; the Clark Class II

The fissurotomy class I, Cala Lilly Class I and the Clark Class II are fairly radical departures from GB Blacks system of preparing and restoring posterior teeth. These new cavity designs are based on adhesive composite restorative materials and engineered to resists tooth fracturing (Figs. 8A-C). The new primary goal of first-time interproximal caries restoration is to avoid connecting the occusal to the interproximal, a concept that Simonson first advocated. The different sizes of the occlusal portion of the new Clark Class II cavity preparation are summarized in Table 2.

A New Fissurotomy Technique (Occlusal Portion):

This new technique has five important components. First, the concept of sealing over caries and grossly contaminated pits and fissures is questioned, and replaced by exacting micromechanical instrumentation. Second, the size and shape of fissure preparation burs is completely modernized with the development of the Fissurotomy Bur System (SS White, Lakewood, NJ). Figure 9 contrasts the precision-engineered SS white Fissurotomy NTF Bur tip size with standard #556 fissure bur. The #556 is unfortunately the most utilized operative bur in dentistry and is largely responsible for the current epidemic of cracked teeth. Today my protocol involves the use of both the Fissurotomy Original Bur as well as the narrower Fissurotomy NTF Bur (Fig. 10). Fissurotomy Burs are scientifically developed instruments for the diagnosis and treatment of hidden caries and should be utilized to create proper preparation form and function for the placement of composite restorations. The Fissurotomy NTF Bur is ideal for ultraconservative micro preparations of pit and fissure defects. The thin carbide tip of the Fissurotomy Burs will not strip quickly like thin diamonds. Third, each occlusal defect is addressed separately, wherein the clinician should avoid the temptation to connect the dots. Fourth, the restorative material of choice is a robust, filled composite such as a flowable composite and/or heated paste composite. Fifth, the use of advanced clinical magnification ranging from 3.5X to 16X is imperative.

The Calla Lilly (occlusal portion): The Cala Lilly, a beautiful trumpet shaped flower (Fig. 11). I use its name to describe the new cavity shape for medium to large sized Class I composites. TradiFigure tional parallel walled cavity preparations have not been shown to provide the adequate volume of enamel rod engagement. Compounding the problem is parallel cavity walls that do not afford proper angle of intersection of enamel rods to provide long term splinting of posterior tooth.

The Clark Class II (Interproximal Portion)

The goal of first-time interproxi- mal caries restoration is to avoid connecting the occlusal to the in- terproximal, which is a concept that Simonson first advocated. The next evolution of this design to be determined is the saucer shape with serpen- tine/disappearing margins. The final change is discarding and re- placing old filling techniques, ma- trixing systems and curing techniques. 

Figures 12-19 demonstrate a clinical example of a combined Cala Lilly occlusal and Clark Class II interproximal. The beau- tiful first bicuspid in figure 12 is from a twenty-eight year old male. It shows an early incomplete coro- nal fracture, based on the magni- fied view of the mesial marginal ridge and according to the guide that we published in the Journal of Esthetic and Restorative Den- tistry and most recently in Dentistry Today. Earlier in my practice, I would have turned this Class I (occlusal) amalgam one into a Class II or MO composite or amalgam, because I would not have seen the fracture undermin- ing the buccal cusp that is not visible at less than 10X. In addi- tion, I suspected but did not un- derstand that turning this 

Class I into a traditional Class II with a mesial box form would further weaken this already iatrogenically compromised tooth. There is now a better approach, one that does not necessarily in- volve an indirect procedure such as a crown or onlay.  

23.png 22.png 21.png

Figures 21-23Low medium and high magnification views of the Shape and Shine diamond impregnated polisher in a low speed, latch grip handpiece. The thin cup allows the polisher to be flanged deeply into interproximal areas. The physics of the cup seem to provide better control than the point polishers. Therefore this has become the go to polisher to achieve the breathtaking finish that we are finally seeing with modern composite dentistry.

As we continue with the bicus- pid in image 12, the occlusal is treated first, and a Calla Lilly shape that engages the bulk of the occlusal table is prepared and re- stored (Fig. 13). The Calla Lilly Class 1 will be explored in future articles. The interproximal is then addressed separately to simplify the process and to control C factor. Note the saucer shape on the me- sial (Fig. 14). This flattened cavity shape requires a completely new filling protocol and peripherals. Instead of metal sectional matrix bands, wedges and separators, we utilize transparent anatomic sec- tional matrix bands (Bioclear Matrix System), translucent Interproximators and a single load technique with an injection molded process where resin, flow- able composite and then paste are loaded in sequence without stop- ping to light cure the individual components. The restoration is light cured with one or multiple curing lights from occlusal, buccal and lingual with this fully trans- lucent system. The result is a seamless, rounded restoration that delivers breathtaking results (Fig. 15-17). Better is rarely faster in dentistry, but our test clinicians report both (better/faster) once they are past the learning curve. 

One year post-operative view of the case is compared to the im- mediate post-operative view in figures 18 and 19. SEM analysis of in vivo wear is showing that the infinity edge margin has superior wear and stain resistance when compared with G.V. Black prepa- ration, with or without a bevel. Interesting, the infinity edge margin of the Cala Lilly occlusal preparation and the infinity edge to be determined margin of the Clark Class II sau- cer preparation both are showing better marginal integrity than that of porcelain inlays. In both comparisons, it is the resistance to marginal ditching of the infin- ity edge margin that appears to provide the advantage.

an These Things Last?

Early posterior composites showed unacceptable wear. Microfills like Heliomolar had excellent wear re- sistance but mediocre strength. Marginal ridge fracture was com- mon. Many modern composites now exhibit excellent strength and wear resistance. In several studies, composite/enamel bond- ing has exhibited very lengthy in vitro success that does not deteri- orate over time.7 The key is that the initial bond must be exquisite and engage large areas of enamel, such as seen in enamel-based por- celain and composite veneers.

In another unpublished study, CRA scientists assisted me for an in office recall study. We docu- mented patients with minimally traumatic Class I composite resto- rations that had been in service for three to seven years. In 107 poste- rior teeth, 100% of the composites were retained. Excess wear was present in some samples that uti- lized flowable composite alone. The combination paste/flowable cases showed the best wear resistance in SEM evaluation (Fig. 20) and slight staining was present in less than 5% of all samples. 


The modern composite restora- tions described in this article are designed to be minimally trau- matic, maximally esthetic, wear resistant and strong. The final leg of the journey is finishing and pol- ishing. One of my favorite diamond impregnated polishers, the Shape and Shine by Clinical Research Dental (Figs. 21-23) provides the best control and shape for creating an unbelievable finish. Care should be taken in polishing not to create excessive heat, which actually cooks the composite and leaves a dull, weakened surface. One of the most important components of my composite hands on courses is the shaping and polishing exercise. Most of us have never been adequately trained to achieve ideal composite finishes, and 90% of composite finishing systems sold today are either redundant or dif- ficult to use.


As we introduce the concepts sum- marized in this article to practic- ing dentists, they show a broad spectrum of responses from shock and disbelief to sheer exuberance. As these cavity shapes are imple- mented we will see a dramatic re- duction in the rate of tooth frac- turing. We also anticipate that these restorations will outlast the Class II amalgams that have served so well in the past. This very brief article can only touch on the dramatic differences of this minimally traumatic and incredi- bly durable direct composite.A full instructional DVD, together with a textbook and hands on tellectual property of the author patent pending and may be used solely with permission.


Dr. David Clark has financial in- terest in several of the products mentioned in this article. The Bioclear Matrix System, hands on courses, training DVDs and fis- surotomy burs and polishers are available through Clinical Re- search Dental at www.clinicalrearchdental.com  and  www.bio-clearmatrix.com/

Bio?? Oral Health welcomes this original article.


1. DeRouen TA, et al. Neurobehavioral effects of dental amalgam in children: a randomized clinical trial. JAMA. 2006 Apr 19; 295(15): 1784-92.

2. Van Nieuwenhuysen JP, DHoore W, Carvalho J, Qvist V. Long-term evaluation of extensive restora- tions in permanent teeth. J Dent. 2003; 31: 395-405.

3. Sjogren P, Halling A. Survival time of class II molar restorations in relation to patient and dental health insurance costs for treatment. Swed Dent J. 2002; 26: 59-66.

4. Mjor IA, Dahl JE, Moorhead JE. Placement and replacement of restorations in primary teeth. Acta Odontol Scand. 2002; 60: 25-28.

5. Clark DJ, Paquette JM, Sheets CJ, Definitive diag- nosis of early enamel and dentinal cracks based on microscopic evaluation. J Esthet Restor Dent 2003; 15 (spec issue): 7-17.

6. Clark DJ. The Epidemic of cracked and fracturing teeth. Dent Today 2007 July: 48-51.

7. Swift EJ Jr, Friedman MJ. J Esthet Restor Dent. 2006; 18(2): 110-3.