Extension for Prevention: Margin Placement

Extension for Prevention: Margin Placement

Thomas D. Larson, D.D.S., M.S.D.*:
The following article is the third and final installment in the current series** Dr. Larson has contributed to Northwest Dentistry in his, and our, continuing commitment to presenting topics which address clinical situations and choices our dentist/readers deal with on a daily basis. “I chose Northwest Dentistry,” Dr. Larson told us, “because if I were to publish in national  journals, many of my former students would never see the articles. They are the audience to which my academic career is devoted. They are the audience that I am trying to affect by  publishing best practices.” The Publications Committee feels that to have someone of Dr. Larson’s academic experience and credentials providing our journal with his extensive yet  concise reviews of the literature is of great value, and we have dedicated a portion of our clinical department to these articles.

This article will review the concept of extension for prevention popularized by G.V. Black around the early 1900s. Concepts of extension and prevention have changed over the years with  a more informed knowledge of the caries process, improved materials, cutting  instruments, and techniques. The reasons for placement of the outline form relative to the tooth  morphology, gingival tissue, relationship to adjacent teeth, and the choice of material will be described for all of the materials used in restorative dentistry. Research will be cited 
to support the scientific basis for outline form placement.
Extension for prevention as a concept came about by studying where plaque would form on teeth and where the contours of teeth and the assumed natural protection of the gingival tissues prevented plaque build up. Further study as to where caries formed on teeth on occlusal surfaces, proximal surfaces, and axial surfaces (facial and lingual) caused G.V. Black to  formulate a plan of restoration that would avoid “at risk” surfaces of teeth.1,2 The cavity design also took into consideration the strength of the material being used and its retention in the tooth. Restoration of teeth was accomplished using amalgam (reformulated by Black in 1895), gold (gold leaf shown to be cohesive in 1855 by heating), silicate cement3,4 (introduced in 1908), and gutta percha (found in 1842). Tin foil was used in lieu of gold leaf from 1873 to at least 1893. 
At the time, there was not a clear understanding of how caries progressed, nor that it could be reversed. It was seen as an acidic process, specifically due to lactic acid, that caused  cavities in teeth. The spread of caries caused cavitation of the enamel, spread along the dentoenamel junction with penetration of bacteria into dentinal tubules. Subsequent  dissolution of the calcium structure of dentin occurs as a result of the acid dissolution. There was some knowledge of several bacteria such as streptococcus infecting the lesions.1 
Given the knowledge of that time as it was first introduced in a manuscript in 1891 in Dental Cosmos (the forerunner of the Journal of the American Dental Association) and later in a  textbook in 1908, the extension of cavity preparations to limit recurrence of caries was a logical outcome. This meant that on occlusal surfaces, all outline forms were extended to smooth surfaces, including all grooves or pits in the outline; proximal extensions were made to the facial and lingual line angles of the teeth to an area of tooth “cleansed” by the  action of chewing food; gingival margins were extended beneath the gingival tissue to protect them from plaque; and axial cavities on facial and lingual surfaces were extended to the  anatomic line angles of the teeth (mesially and distally), beneath the gingival tissue and occlusal to the crest of convexity, to areas of the tooth that would be naturally cleansed. All  of these extensions were meant to avoid placing the margin of the restoration in areas where plaque would normally accumulate. The extension of the cavity occlusal outline caused  the width of the isthmus to be one third of the intercuspal distance at a minimum, and in all cases at least into the dentinal structure pulpally (also a requirement of resistance form).  These extensions resulted in a massive loss of often sound tooth structure, all to prevent future cavities. 
At this time in history, tooth brushing was neither routine nor common. By 1914, dental hygienists were used in the school systems to teach mouth hygiene, and tooth brushing  became more important.5  Fluoride was in some water supplies naturally, but was not known about as a preventive agent. Dentists recognized some patients’ immunity to caries, but  did not understand why they were immune. As soon as decalcification occurred on the tooth, it was recommended the tooth be restored, as no concept of recalcification was understood.6  
Over the years, outline forms of direct materials have been modified from these original concepts of extension for prevention. Prime7 (1928), Bronner8 (1931), and Markley(1951) all suggested modifications to the original extensions for amalgam preparations advocated by Black. Markley originally advocated narrowing the isthmus width to no more than  one quarter of the intercuspal dimension and narrowing the proximal extensions at the level of the marginal ridge. In later years, he designed the 330 bur with rounded flutes to give  the internal line angles of the preparation a rounded rather than sharp feature (Figure 1). In the 1960s and 1970s, the outline form for amalgam restoration was made smaller, with  extension on the proximal surfaces to clear the adjacent tooth by 0.5mm and for the undercut of the proximal walls to decrease to about a three degrees undercut (on the working  cusp side) to proximal walls parallel to the long axis (on the non-working cusp side).9 The proximal extensions are to clear the adjacent tooth by 0.5 mm. In the 1960s and 1970s,  outline form for composite restorations became smaller and smaller as we began to bond to the enamel first and later to dentin. Gingival extensions of proximal boxes have remained at  0.5mm from the adjacent tooth for both composite and amalgam outline forms since the 1970s (Figure 2).d Lesionspecific designs for various materials such as preventive resin  restorations began to emerge in the late1900s. Lesion-specific outlines call for extension adequate for convenience and resistance form, and may include bonding to enamel and dentin  for retention form. Lesion-specific implies no “ideal”, set design of a cavity preparation, but rather is based on the size and position of the caries lesion with a preparation design that  allows visibility and access. 

Figure 1: Examples of outline forms for a conservative amalgam restoration from the 1970s as advocated by Markley on maxillary fi rst premolar MO and mandibular second premolar DO.   Figure 2: Typical outline form for amalgam restorations from the 1960s and 1970s.

Cavity Design, Outline Form
Modern concepts of the extension of outline forms for direct and indirect restorations rely on more scientifically defined criteria. In the past, observation and empirical study modified  by a less sophisticated understanding of the sciences helped the pioneers of operative dentistry to formulate restoration concepts. Though modified by degree, G.V. Black’s original  teaching of the seven steps of cavity preparation still holds.6 Those steps are:
1. Outline form
2. Resistance form
3. Retention form
4. Convenience form
5. Removal of remaining caries
6. Finish of the enamel wall
7. Cleansing of the cavity
Modern science has altered tooth restoration outline form in significant ways. There is a call in the literature for minimally invasive preparations, though that phase has numerous and  sometimes erroneous meanings, depending on author. 
Decalcification/Recalcification of Enamel 
Dental caries is a diet-supported disease. The Vipeholm study10 in the 1950s dramatically demonstrated the role diet plays in the etiology of rampant caries. Children who consumed  sweets frequently during the day developed aggressive dental caries patterns, whereas children who consumed the same amount of sweets but confined them to the three meals did  not develop such decay. Similarly, sticky sweets like toffees and caramels were more cariogenic than less retentive forms. 
Teeth demineralize and remineralize based on the pH and length of exposure to a pH of 4.5-5.5. This was first proposed in 1949.11 Below that pH, enamel is demineralized; above that  pH hydroxyapatite is strong enough to prevent demineralization. Remineralization is improved in the presence of fluoride and sufficient calcium phosphates normally found in saliva.12,13,14 Because of these concepts, extension of a tooth preparation need not include decalcified areas of enamel if they are confined to superficial enamel. Fluoride found in  many water supplies and in many types of toothpaste is sufficient to remineralize after acid attack.15 Many other therapeutic approaches can be used to prevent caries formation  even in dry mouth conditions. 
Remineralization therapies utilizing fluorides and exogenous calcium phosphates are effective at hardening the enamel of previously demineralized surfaces.17,18,19,20 This should be  the initial therapy, rather than surgical intervention, for incipient lesions on smooth surfaces, together with oral hygiene instruction and dietary modification. Other therapies intending  to alter the microflora of an infected mouth can be used — e.g., xylitol gum and chlorhexidine.16 Because of these remineralization opportunities available daily, outline forms can be significantly smaller. 

Figure 3: Lesion specific cavity design for a Class II composite restoration. The preparation design does not break proximal contact on the facial wall, but the margins clear the  proximal contact on the gingival and lingual walls. This same design can be used for amalgam restorations with 90° cavosurface margins and line angle retention grooves. Photo by Dr.  Ignatius Lee.
Figure 4: Restoration of the DO composite preparation utilizes composite resin to restore interproximal contour and sealants on the remaining occlusal grooves of the premolars and the  molars. Photo by Dr. Ignatius Lee.
Figure 5: Completed restoration with occlusal seaalants and a Class II composite resin on the distal of the second premolar. Photo by Dr. Ignatius Lee. 

Tooth Bonding and Sealants
Bonding to enamel and dentin provides some significant benefits to restorations of all types, generally improving durability. Specifically, tooth bonding decreases microleakage,21,22  diminishes postoperative sensitivity, improves resistance to cuspal fracture and flexure,24,25,26,27 improves retention strength,28,29,30 and improves marginal stability of amalgam.31 
Secondary caries is most often a reason for restoration replacement. Eighty to ninety percent of margins with secondary caries are found at the gingival margin.32 Marginal gaps most  often are the cause of secondary caries.33,34,35,36,37,38 Marginal preparation is probably a key to prevention of secondary caries, as well as the use of bonding, regardless of  material used for interproximal restoration.39 Use of hand instruments to plane the gingival wall and remove loose enamel rods improves marginal adaptation, especially in smaller cavity  forms. 
Pits and fissures are best managed in tooth preparation using pit and fissure sealants when caries may be suspected or for prevention.40 Evidence-based recommendations40 are to  use composite resin restoration in combination with sealants for initial lesions to minimize tooth reduction and outline form. This combination is termed preventive resin restoration  (PPR).41,42 Composite resin is most adaptable in these circumstances because it does not require placement into dentin as amalgam restorations do (resistance form). The cavity form  is lesion-specific, with size and extent dictated by the presence of caries and sufficient convenience form to insert and cure the material. There is no “ideal” outline form for either composite or amalgam, as each preparation on the occlusal surfaces is lesion-specific. 
Pit and fissure sealants may also be used with occlusal amalgam restorations when the lesion extends more deeply into dentin, to limit outline form. When used in combination with  sealants, this is the most durable restoration,31 requiring the least removal of tooth structure, maintaining the strongest tooth. In the ten-year clinical study cited,31 the bonded  amalgam with sealants showed superior clinical performance and longevity compared to unbonded amalgam. In a similar nine-year study, the bonded and sealed lesion-specific occlusal  amalgam outperformed the unbonded conventional occlusal outline form amalgam.43 An additional five-year clinical study shows the bonded and unbonded amalgam had no difference  in marginal integrity with conventional tooth preparation, but the bonded amalgam had excellent retention and marginal integrity in unretentive cavity forms.44 These different studies  show that after a five-year period, marginal breakdown for unbonded amalgams increases. 
Bonding can very significantly decrease outline form. Consider the Class II amalgam preparation (Figures 3, 4, 5). If the restoration of a virgin lesion will be bonded, the facial extension  (much like that for a Class II posterior composite) need not be extended all the way through the proximal contact facially. The preparation need only remove loose enamel and the  carious lesion. Gingival margins must still extend about 0.5 mm from the adjacent tooth (i.e., they must clear the contact, as caries occurs most often lingual and gingival to the  proximal contact).45 Though requiring hand instruments to finish all enamel margins, this much smaller outline form results in a considerably stronger tooth, less likely to fracture. In  fact, if desirable, the occlusal extension of this outline could be managed with a pit and fissure sealant with the use of a proximal box-only. The material still requires convergent walls  occlusally and proximal undercuts (line angle retention grooves)46 together with the bonding to be retentive enough with adequate resistance form to support occlusion. This type of  restoration requires very careful condensation of the amalgam in order to achieve an adequate marginal finish neither under-filled, nor overfilled.47 This is an outline form far different  from what many Class II restorations look like today.  
Effect of Tooth Preparation on the Strength of Teeth
Tooth preparation leads to a loss of tooth strength, with increases in the amount of tooth structure removal causing greater weakness.48,49 In these and further studies, increasing  the isthmus width from one quarter to one third of the intercuspal distance (measured from facial to lingual cusp tip) significantly decreases tooth strength. Even preparations cut with  MOD cavities have no significant strength difference between occlusal cavities cut with the same isthmus width.50 However, the fracture pattern of the larger MOD cavities invariably  leads to a more catastrophic fracture compared to cavities such as box-only Class II preparations.51 More conservative amalgam restorations exhibit fewer breakdowns than larger restorations.52 
Fracture toughness of teeth is related to propagation of dentinal tears caused by tooth preparation and by cyclic loading of the teeth from activities such as bruxing. The behavior of  dentin in testing suggests a distribution of pre-existing defects in dentin and that normal mastication and bruxing can initiate dentinal cracks, especially when the normal anatomy of  the tooth has been altered by tooth preparation.53,54 Fatigue fractures are caused by more than successive static force applications, but instead by a cyclic fatigue mechanism that involves crack-tip blunting and resharpening as a result of the cyclic application of force.55,56
Any minimization of tooth preparation size and extension of Class I or Class II preparations will increase the longevity of both the restoration and the tooth, and will limit tooth fracture  because it creates fewer and smaller dentinal tears, located more centrally rather than facial or lingual.57,58 
In an interesting in vitro study, molar cusps were intentionally undermined leaving no dentin support and then restored with unbonded amalgam, bonded amalgam, and bonded  composite. The cusps were fractured with an Instron machine. The force required to fracture the undermined but restored cusps was intermediate between cusps with no support and  cusps supported by dentin. There were significant differences between the unbonded and bonded specimens. There was a significant difference between the restored versus dentin- supported cusps and between the dentin-supported and unsupported cusps.58 While unsupported enamel is made stronger by bonding, it is still weaker than enamel supported by  dentin.59 
Restorative Material Improvements 
The key to decreasing the development of secondary or marginal caries is to maintain marginal integrity of the materials. Secondary caries is thought to be associated with marginal gaps, voids, and openings along the margins of existing restorations.60,61 As stated previously, bonding of materials will generally improve their durability and marginal integrity.31  Studies of restoration longevity generally indicate that indirect materials outlast direct materials. Two meta reviews of clinical studies found the most common reasons for restoration  replacement were secondary caries, fracture, marginal discrepancies, wear, and postoperative sensitivity.62,63
Differences in materials can be significant, as recent research has shown that composites have a higher rate of secondary caries the further posteriorly they are placed compared to  amalgam.64 The rate of secondary caries of composites in clinical studies has been shown to be 3.5 times that of amalgam restorations.65 From these and other studies it is apparent  that the choice of material for specific restoration locations becomes critical to their longevity. 
Amalgam as material has significantly improved since the days of Dr. Black. High copper amalgams, popularized in the late 1960s/early 1970s, have become the standard amalgam  materials used. High copper amalgam decreases the gamma two phase, increasing edge strength and decreasing creep. Both properties cause significant issues with marginal  breakdown, leading to more secondary caries. High copper amalgams exhibit fewer breakdowns, with dispersed-phase alloys superior to spherical particle alloys.57 Zinc as an additive (approximately one percent by weight) improves the fatigue resistance and results in better marginal adaptation with superior resistance to marginal breakdown.66 

Figure 6: Top of picture is enamel with fracture of loose enamel on an etched and bonded enamel margin, leaving an open margin, still attached to composite material at the bottom half of the picture. Secondary caries most often occurs at marginal gaps. Co

Gingival Extensions
Black and others believed the extension of a margin sub-gingivally imparted some protection from plaque and caries formation.2 As research on caries and periodontal disease has  improved, today we know this assumption was incorrect. Periodontal research is conclusive in finding that placement of subgingival margins results in loss of attachment in a one- to  three-year time frame and is detrimental to the periodontal health.67 An in vivo study over a 26-year period demonstrates subgingival margins do not protect gingival health.  Supragingival compared to subgingival margins improve gingival health.68 Bacteriological studies from various patient populations show convincingly that subgingival locations impart no lack of bacteria that cause caries, both smooth surface caries and root caries.69,70,71 Extension of gingival margins subgingivally is only necessary to include caries or old  restoratives, to create greater length of preparation for retention, or for esthetics. 
Composite Restorations 
Composite restorations generally are lesion specific, regardless of their location in the mouth. That is, the outline form is created with adequate convenience form to excavate carious tissue, establish a sound perimeter, and insert and polish the material. Bonding of these restorations has lengthened their durability. As noted earlier, in Class II situations, the further  posterior the composite, the more likely it is that secondary caries will form. This can be attributed to several significant factors.72 The surface free energy of composite materials  attracts the Gram-negative bacteria that cause caries. The gingival margin of the preparation must be extended 0.5 mm from the adjacent tooth and must be properly finished. If the  operator does not use a hand instrument to remove loose enamel rods, the loose enamel will be fractured due to the polymerization contraction of the composite resin while setting,  leaving a significant marginal gap (Figure 6). Marginal gaps cause secondary caries. The gingival margin is the most difficult to isolate properly. Proper isolation is a key to good  adaptation and bonding, as is adequate condensation force with no pull back of the material by the condenser. 
As explained earlier, Class I outline forms are lesion-specific, as are Class II, III, IV, V, and VI, all limited to removal of caries and convenience for insertion. The preparation outline  must extend to an area of the tooth where enamel is sound, not friable. Caries must be removed and sufficient convenience form allowed for insertion. Blind margins in contact with  adjacent teeth are adequate as long as they can be planed to remove loose enamel and can be bonded. Leaving undermined enamel is risky, as its strength is between that of  unbonded enamel and fully supported enamel.58 
Glass Ionomer 
The extensions of tooth preparations for glass ionomer are the same as for composite materials, but are applicable to generally Class I, III, V due to the weakness of the material. The  outline form is lesion-specific, made to accommodate adequate convenience form for insertion. As this material bonds to tooth structure and provides therapeutic release of fluoride, it  need not be extended beyond the lesionspecific criteria. Occasionally glass ionomer materials are used for ART (Atruamatic Restorative Treatment) for the temporary restoration of carious teeth, and are used in Class II, IV, and VI as well. Glass ionomer is also used as a temporary material for use in rampant caries patients.16 
Indirect Materials 
The efficacy of bonding has improved our ability to fit the indirect materials with significant durability. Generally used for broken down teeth, for the replacement of large restorations,  or to correct some esthetic or functional concern, indirect restorations, including cast gold, porcelain, and indirect laboratory processed composite, can utilize site-specific preparation  outlines required to correct the variety of these problems. 
Supra-gingival margins generally provide for better gingival health over long periods of time.67 On many occasions porcelain margins can be bonded at or within 0.5 mm of the gingival  crest with very little effect on the health of the gingival tissues. Extending sub-gingivally should be done for either significant esthetic concerns or for length necessary for retention.  The most favorable response from the gingiva is achieved with the margins at or above the crest of the gingival tissue.73,74 
Extensions of partial-coverage indirect restorations should allow for the direct finishing of the margins to permit the best marginal fit with the smallest gap possible. As gap sizes  increase, secondary caries increases.32,33,34,35,38,39 
Occlusal forces can cause breakdown of margins if the margin is exposed to direct occlusal pressure.75,76,77,78 Most commonly, margins fracture, leaving gaps that promote  secondary caries. With cast gold, margins that undergo direct occlusal pressure may also open due to the force of occlusion, especially from bruxing, leaving a gap that can become  carious. Extension of preparation margins for indirect materials to avoid direct occlusal contact in either maximum intercuspation or lateral movement will improve durability. 
Improved materials and advancing the understanding of science has led to lesion-specific and site-specific outline forms for all restorative materials. While vastly different by degree  from the outline forms described by G.V. Black in the early 1900s, the modern designs are determined by scientific principles following the principles of cavity preparation outlined  originally by G.V. Black.

1. Black GV. The management of enamel margins. Dental Cosmos 1891;33:85-100.
2. Black GV. A Work on Operative Dentistry, First Edition. Volume I. Chicago: Medico-Dental Publishing Co., 1908, pages 60-73.
3. Volker CJ. Dental silicate cements in theory and practice. Dental Cosmos 1916;68:1,098.
4. Glenner RA. Dental cements and tooth colored fi lling materials. Bull Hist Dent 1993;41(3): 111-15.
5. Black AD. GV Black’s Work on Operative Dentistry, Seventh Edition, Volume 3. Treatment of Dental Caries. Chicago: Medico-Dental Publishing Co., 1936, pages 12-42.
6. Black GV. A Work on Operative Dentistry, First Edition, Volume II. Chicago: Medico-Dental Publishing Co., 1908, pages 105-116.
7. Prime JM. A plea for conservatism in operative procedures. JADA 1928;15:1,234.
8. Bronner FJ. Dental Cosmos 1931;73:577.
9. Markley MR. Restorations of silver amalgam. JADA 1951 Aug;43(2):133-46.
10. Gustafsson BE, Quensel CE, Lanke LS, Lundqvist C, Grahnen H, Bonow BE, Krasse B. The Vipeholm dental caries study; the effect of different levels of carbohydrate intake on  caries activity in 436 individuals observed for five years. Acta Odontol Scand 1954 Sep;11(3-4):232-64.
11. Stephan RM. Intra-oral hydrogen ion concentrations associated with dental caries activity. J Dent Res 1944;23:257-266.
12. Featherstone JD. Remineralization, the natural caries repair process - the need for new approaches. Adv Dent Res 2009;21(1):4-7.
13. ten Cate JM, Buijs MJ, Damen JJ. pH-cycling of enamel and dentin lesions in the presence of low concentrations of fl uoride. Eur J Oral Sci 1995 Dec;103(6):362-7.
14. Casals E, Boukpessi T, McQueen CM, Eversole SL, Faller RV. Anticaries potential of commercial dentifrices as determined by fluoridation and remineralization effi ciency. J Contemp  Dent Pract 2007 Nov 1;8(7):1-10.
15. Casals E, Boukpessi T, McQueen CM, Eversole SL, Faller RV. Anticaries potential of commercial dentifrices as determined by fl uoridation and remineralization effi ciency. J Contemp  Dent Pract 2007 Nov 1;8(7):1-10.
16. Hildebrandt G, Larson TD. Management of rampant caries. Northwest Dent Jan-Feb 2009;88(1):35-45,51.
17. Reynolds EC. Remineralization of enamel subsurface lesions by casein phosphopeptidestabilized calcium phosphate solutions. J Dent Res 1997 Sep;76(9):1,587-95.
18. Hay KD, Thomson WM. A clinical trial of the anticaries effi cacy of casein derivatives complexed with calcium phosphate in patients with salivary gland dysfunction. Oral Surg Oral 
Med Oral Pathol Oral Radiol Endod 2002 Mar;93(3):271-5.
19. Shen P, Cai F, Nowicki A, Vincent J, Reynolds EC. Remineralization of enamel subsurface lesions by sugar-free chewing gum containing casein phosphopeptide-amorphous calcium  phosphate. J Dent Res 2001 Dec;80(12):2,066-70.
20. Iijima Y, Cai F, Shen P, Walker G, Reynolds C, Reynolds EC. Acid resistance of enamel subsurface lesions remineralized by a sugar-free chewing gum containing casein  phosphopeptideamorphous calcium phosphate. Caries Res 2004 Nov-Dec;38(6):551-6.
21. Larson TD. The clinical signifi cance and management of microleakage. Part one. Northwest Dent 2005 Jan-Feb;84(1):23-5,28-9,31.
22. Larson TD. The clinical signifi cance and management of microleakage. Part two. Northwest Dent 2005 Mar-Apr;84(2):15-9. 
23. Murray PE, Hafez AA, Smith AJ, Cox CF. Bacterial microleakage and pulp infl ammation associated with various restorative materials. Dent Mater 2002 Sep 18(6):470-8. 
24. Boyer DB, Roth L. Fracture resistance of teeth with bonded amalgams. Am J Dent 1994 Apr;7(2):91-4.
25. Pilo R, Brosh T, Chweiden H. Cusp reinforcement by bonding of amalgam restorations. J Dent 1998 Jul- Aug;26(5-6):467-72.
26. Setcos JC, Staninec M, Wilson NH. Bonding of amalgam restorations: existing knowledge and future prospects. Oper Dent 2000 Mar- Apr;25(2):121-9.
27. Zidan O, Abdel-Keriem. The effect of amalgam bonding on the stiffness of teeth weakened by cavity preparation. Dent Mater 2003 Nov;19(7):680-5.
28. Grobler SR, Oberholzer TG, Rossouw RJ, Grobler-Rabie A, Van Wyk Kotze TJ. Shear bond strength, microleakage, and confocal studies of 4 amalgam alloy bonding agents.  Quintessence Int 2000 Jul-Aug;31(7):501-8.
29. Neme AL, Evans DB, Maxson BB. Evaluation of dental adhesive systems with amalgam and resin composite restorations: comparison of microleakage and bond strength results. Oper Dent Nov-Dec 2000;25(6):512-9. 
30. Cooley RL, Tseng EY, Barkmeier WW. Dentinal bond strengths and microleakage of a 4-META adhesive to amalgam and composite resin. Quintessence Int 1991 Dec;22(12):979-83. 
31. Mertz-Fairhurst EJ, Curtis Jr. JW, Ergle JW, Rueggeberg, FA, Adair SM. Ultraconservative and cariostatic sealed restorations: Results at year 10 years. JADA 1998 Jan;129(1):55- 66.
32. Mjör IA. The location of clinically diagnosed secondary caries. Quintessence Int 1998 May;29(5):313-7.
33. Kidd EA, Beighton D. Prediction of secondary caries around tooth-colored restorations: a clinical and microbiological study. J Dent Res 1996 Dec;75(12):1,942-6.
34. Rezwani-Kaminski T, Kamann W, Gaengler P. Secondary caries susceptibility of teeth with long-term performing composite restorations. J Oral Rehabil 2002 Dec;29(12):1,131-8.
35. Kidd EA, Joyston-Bechal S, Beighton D. Marginal ditching and staining as a predictor of secondary caries around amalgam restorations: a clinical and microbiological study. J Dent  Res 1995 May;74(5):1,206-11.
36. Kidd EA, O’Hara JW. The caries status of occlusal amalgam restorations with marginal defects. J Dent Res 1990 Jun;69(6):1,275-7.
37. Mjör IA, Toffenetti F. Secondary caries: a literature review with case reports. Quintessence Int 2000 Mar;31(3):151. 
38. Hodges DJ, Mangum FI, Ward MT. Relationship between gap width and recurrent dental caries beneath occlusal margins of amalgam restorations. Community Dent Oral Epidemiol 1995 Aug;23(4):200-4. 
39. Larson TD. The Clinical Signifi cance of Marginal Fit. Northwest Dent 2012 Jan-Feb;91(1):22-29. 
40. Beauchamp J, Caufi eld PW, Crall JJ, Donly K, Feigal R, Gooch B, Ismail A, Kohn W, Siegal M, Simonsen R. Evidence-based clinical recommendations for the use of pit-and-fissure  
sealants: a report of the American Dental Association Council on Scientific Affairs. American Dental Association Council on Scientifi c Affairs. JADA 2008 Mar;139(3):257-68. 
41. Simonsen RJ. Preventive resin restoration. Innovative use of sealants in restorative dentistry. Clin Prev Dent 1982 Jul-Aug;4(4):27-9. 
42. Simonsen RJ. Preventive resin restorations and sealants in light of current evidence. Dent Clin North Am 2005 Oct;49(4):815-23, vii.
43. Mertz-Fairhurst EJ, Adair SM, Sams DR, Curtis JW Jr, Ergle JW, Hawkins KI, Mackert JR Jr, O’Dell NL, Richards EE, Rueggeberg F et al. Cariostatic and ultraconservative sealed restorations: nine-year results among children and adults. ASDC J Dent Child 1995 Mar- Apr;62(2):97-107.
44. Mach Z, Regent J, Staninec M, Mrklas L, Setcos JC. The integrity of bonded amalgam restorations: a clinical evaluation after fi ve years. JADA 2002 Apr; 133(4):460-7. 
45. Otto PF, Rule JT. Relationship between proximal cavity design and recurrent caries. JADA 1988 Jun;116(7):867-70. 
46. Summitt JB, Howell ML, Burgess JO, Dutton FB, Osborne JW. Effect of grooves on resistance form of conservative Class 2 amalgams. Oper Dent 1992 Mar-Apr;17(2):50-6.
47. Duncalf WV, Wilson NH. A comparison of the marginal and internal adaptation of amalgam and resin composite restorations in small to moderate-sized Class II preparations of conventional design. Quintessence Int 2000 May;31(5):347-52.
48. Vale WA. Cavity preparation. Irish Dental Review 1956;2:33. 
49. Vale WA. Cavity preparation and further thoughts on high speed. B Dent J 1959; 107:333. 
50. Larson TD, Douglas WH, Geistfeld RE. Effect of cavity preparation on strength of unrestored teeth. Oper Dent 1981;6(1):2-5. 
51. el-Mowafy OM. Fracture strength and fracture patterns of maxillary premolars with approximal slot cavities. Oper Dent 1993 Jul- Aug;18(4):160-6.
52. Osborne JW, Gale EN. Failure at the margin of amalgams as affected by cavity width, tooth position, and alloy selection. J Dent Res Mar 1981;60(3):681-685.
53. Staninec M, Marshall GW, Hilton JF, et al. Ultimate tensile strength of dentin: evidence for damage mechanics approach to dentin failure. J Biomed Mater Res 2002;63(3):342-5.
54. Rasmussen ST, Patchin RE, Scott DB, Heuer AH. Fracture properties of human enamel and dentin. J Dent Res 1976 Jan-Feb;55(1):154-64. 
55. Kruzic JJ, Nalla RK, Kinney JH, Ritchie RO. Mechanistic aspects of in vitro fatiguecrack growth in dentin. Biomaterials 2005 Apr;26(10):1,195-1,204.
56. Nalla RK, Kinney JH, Marshall SJ, Ritchie RO. On the in vitro fatigue behavior of human dentin: effect of mean stress. J Dent Res 2004 Mar; 83(3):211-15.
57. Osborne JW, Gale EN. Failure at the margin of amalgams as affected by cavity width, tooth position, and alloy selection. J Dent Res 1981 Mar;60(3):681-5.
58. Latino C, Troendle K, Summitt JB. Support of undermined occlusal enamel provided by restorative materials. Quintessence Int 2001 Apr;32(4):287-91.
59. Abu-Hanna AA, Mjör IA. Resin composite reinforcement of undermined enamel. Oper Dent 2004 Mar-Apr;29(2):234-7. 
60. Mjor IA, Toffenetti F. Secondary caries: a literature review with case reports. Quintessence Int 2000 Mar; 31(3): 165-79. 
61. Goldberg AJ. Deterioration of restorative materials and the risk for secondary caries. Adv Dent Res 1990 Jun;4:14-18. 
62. Hickel R, Manhart J. Longevity of restorations in posterior teeth and reasons for failure. J Adhes Dent 2001 Spring;3(1):45-64.
63. Manhart J, Chen H, Hamm G, Hickel R. Buonocore Memorial Lecture. Review of the clinical survival of direct and indirect restorations in posterior teeth of the permanent dentition. Oper Dent 2004 Sep-Oct;29(5):481-508.
64. Kirkevang LL, Vaeth M, Wenzel A. Prevalence and incidence of caries lesions in relation to placement and replacement of fi llings: a longitudinal observational radiographic study of  an adult Danish population. Caries Res 2009;43(4):286-93. Epub 2009 May 8.
65. Bernardo M, Luis H, Martin MD, Leroux BG, Rue T, Leitao J, DeRouen TA. Survival and reasons for failure of amalgam versus composite posterior restorations placed in a randomized clinical trial. JADA 2007 Jun;138(6):775-83. 
66. Watkins JH, Nakajima H, Hanaoka K, Zhao L, Iwamoto T, Okabe T. Effect of zinc on strength and fatigue resistance of amalgam. Dent Mater 1995 Jan;11(1):24-33.
67. Schätzle M, Land NP, Anerud A, Boysen H, Bürgin W, Löe H. The infl uence of margins of restorations of the periodontal tissues over 26 years. J Clin Periodontol 2001 Jan;28(1):57- 64.
68. Gullo CA, Powell RN. The effect of placement of cervical margins of class II amalgam restorations on plaque accumulation and gingival health. J Oral Rehabil 1979 Oct;6(4):317-22. 
69. Gizani S, Papaioannou W, Haffajee AD, Kavvadia K, Quirynen M, Papagiannoulis L. Distribution of selected cariogenic bacteria in fi ve different intra-oral habitats in young children.  Int J Paediatr Dent 2009 May;19(3):193-200. Epub 2008 Dec 14.
70. Mager DL, Ximenez-Fyvie LA, Haffajee AD, Socransky SS. Distribution of selected bacterial species on intraoral surfaces. J Clin Periodontol 2003 Jul;30(7):644-54.
71. Hintao J, Teanpaisan R, Chongsuvivatwong V, Ratarasan C, Dahlen G. The microbiological profi les of saliva, supragingival and subgingival plaque and dental caries in adults with  and without type 2 diabetes mellitus. Oral Microbiol Immunol 2007 Jun;22(3):175-81
72. Larson TD, Why do we polish? Parts one and two. Northwest Dent 2011 May-Jun,Jul- Aug;90(3,4):17-22;31-38.
73. Shaini FJ, Shortall AC, Marquis PM. Clinical performance of porcelain laminate veneers. A retrospective evaluation over a period of 6.5 years. J Oral Rehabil 1997 Aug;24(8):553-9. 
74. Troedson M, Derand T. Effect of margin design, cement polymerization, and angle of loading on stress in porcelain veneers. J Prosthet Dent 1999 Nov;82(5):518-24.
75. Aggarwal V, Logani A, Jain V, Shah N. Effect of cyclic loading on marginal adaptation and bond strength in direct vs. indirect class II MO composite restorations. Oper Dent. 2008  Sep-Oct;33(5):587-92.
76. Cavalcanti AN, Mitsui FH, Silva F, Peris AR, Bedran-Russo A, Marchi GM. Effect of cyclic loading on the bond strength of class II restorations with different composite materials. Oper Dent 2008 Mar-Apr;33(2):163-8.
77. Pongprueksa P, Kuphasuk W, Senawongse P. Effect of elastic cavity wall and occlusal loading on microleakage and dentin bond strength. Oper Dent 2007 Sep-Oct;32(5):466-75.
78. Hoard RJ, Eichmiller FC, Parry EE, Giuseppetti AA. Edge-bevel fracture resistance of three direct-filling materials. Oper Dent 2000 May- Jun;25(3):182-5.
*Dr. Larson is Associate Professor, Department of Restorative Sciences, Division of Operative Dentistry, University of Minnesota School of Dentistry, 8-450 Moos Tower, 515 Delaware  Street S.E., Minneapolis, Minnesota 55455. Email is larso004@umn.edu
**“Why Do We Polish? Part One”, Northwest Dentistry, Volume 90, Number 3, May-June, 2011, pages 17-22; “Why Do We Polish? Part Two”, Northwest Dentistry, Volume 90, Number  4, July- August, 2011, pages 31-38; “The Clinical Signifi cance of Marginal Fit”, Northwest Dentistry, Volume 91, Number 1, January- February 2012, pages 22-29.