- Email: [email protected]
Finishing of amalgam restorations: to what degree is it necessary?
J. Dent.
1989;
55
17: 55-60
Finishing of amalgam restorations: what degree is it necessary?
to
I. W. M. Jeffrey and N. B. Pit&* Department of Conservative Dentistry and * Dental Health Services Research Unit Department of Periodontology and Community Dentistry, Dental School, Dundee, UK KEY WORDS: Amalgam, Operative technique, Review
J. Dent. 1989;
17: 55-60
(Received 26 May 1988;
reviewed 7 July 1988;
accepted 18 July 1988)
ABSTRACT This paper attempts to elicit from the literature evidence to guide the clinician in deciding what degree of finishing and polishing is appropriate for amalgam restorations. Much of the justification for the polishing of amalgams has, in the past, been empirically based. Comprehensive scientific investigation into the validity of this practice does, however, present a number ofmethodological problems which have yet to be successfully overcome. An attempt is made to define what is meant by the term ‘polished’ (in a clinical context) while the benefits of polishing, as far as they are known, are discussed. The roles that cavity preparation, carving and burnishing play in the finishing of a restoration are reviewed, as is the conflicting evidence regarding their respective advantages and disadvantages. The importance of marginal adaptation, the relationship of polishing to secondary caries, and the dilemma in choosing (in appropriate cases) between the replacement of a restoration or its improvement by polishing are also considered.
INTRODUCTION
In operative
dentistry it is commonly regarded as a sine qua non that amalgam restorations, at some time following their placement, should be polished. The justification for the adoption of such a polishing regimen has been based largely on pragmatic beliefs and a priori
arguments, the most obvious of these being the propensity for unpolished aspects of the restoration to encourage plaque formation. This outcome, it is assumed, will lead to the subsequent degradation of the cavity margins by secondary caries which will necessitate replacement of the restoration. In the undergraduate context, and with apparently little variation between the dental schools, there is a place on the student’s record card of work carried out wherein confirmation is given that each restoration inserted has been duly polished. This ‘polishing’ is often specifically graded by supervizing staff. In general practice, however, the state of finish of restorations would appear to be a tripartite matter between the individual beliefs and conscience of the practitioner; the occasional introverted curiosity of the more dentally aware patient; and the surveillance of ‘third-party assessors’ such as a Dental Reference Officer in the UK. In both dental schools and 0 1989 Butterworth 8c Co. Publishers 0300-5712/89/020055-06 $03.00
Ltd.
general practice, however, many dentists apparently regard the absence of a highly polished amalgam surface (elaborately carved to resemble the majority of the occlusal anatomical pits and fissures), or the presence of any slight marginal defect, as ample reason to initiate replacement of the restoration. It must surely be the responsibility of academic dentistry to investigate the fundamental, but routine problem of identifying both the optimal and the acceptable level to which an amalgam restoration should be finished. Investigative studies assessing the need for finishing, determining the effects of (and costs associated with) achieving varying standards of surface polishing and marginal integrity, and ascertaining the sequelae to be expected if certain standards are not met in a representative clinical environment, are all highly desirable. There are, however, significant methodological problems which must be overcome first. The major difficulty concerns the large number of variables to be considered in any such study-the oral hygiene and caries risk of the hosts, the siting and dimension of the particular restorations, the choice and standardization of the finishing techniques employed, the types of material used, the level of skill and representativeness of the operator(s), to name but a few.
56
J. Dent. 1989; 17: No. 2
Very large numbers of patient subjects would have to be employed before proper statistical control could vouchsafe a correct conclusion being made. Another complication is the well-established observation that a great many restorations which are not polished or carved to any great extent exist for long periods without evidence of overt manifestations of harm. The aim of this paper is to examine current opinion with respect to some of the various stages of finishing amalgam restorations and to discuss the appearance of amalgam restorations as a factor influencing the treatment decision to replace them.
POLISHING It is commonly believed to be advisable to polish any material that is to be inserted ‘permanently’ in the mouth because surface roughness predisposes to plaque accumulation and an increased risk of caries; however, polishing has not always been carried out. A simple example is seen in the finishing of the fitting surfaces of metal partial dentures with surfaces contiguous to tooth tissue. No more than a vigorous cleaning and a removal of gross roughnesses from the casting is performed, but in no sense can this be considered true polishing. It is conceded that in this case the denture is removable but it is, nevertheless, eligible to remain in the oral environment for long periods in direct apposition to surfaces susceptible to caries. In addition, the polishing of any material per se in the mouth has been motivated by other factors. These include the presence of sharp, rough or jagged surfaces which could be unpleasant for the patient and which could initiate either acute or chronic damage to the oral mucosa; or food trapping which in turn can alter the surface of metallic restorative materials leading to electrolytically sponsored corrosion and staining. However, the clinical significance of these problems, particularly with respect to the modern high-copper amalgam alloys, in terms of their effect upon the longevity of restorations, is not well understood. One difficulty that arises in any discourse on polishing within the mouth is to define, in exact terms, what is meant by polishing. What, for example, distinguishes a smooth surface from a polished one? It is apparent that, at a technical level, definitive differences are rather subtle and slightly confusing. There is almost a temptation to deal in simplistic terms, such as rough or smooth, until it is realized that these terms are even more confusing, arbitrary and ambiguous. The process of polishing should produce a shiny, mirror-like (when viewed by the naked eye) surface without the production or assistance of supplemental surface lihns created in other situations by the addition of wax or lacquer. In essence, polishing means the formation of a scratch-free surface; the scratches varying from the deep macroscopically evident scores, to fine lines which require a microscope to see.
Only on the final disappearance of all scratches can the surface be said to be truly polished. Even today, the polishing process does not seem to be clearly understood. The usually accepted sequence of events believes that the initial stage is one of abrasion, where the peaks of the score furrows are reduced in height by the action of a series of increasingly fmer abrasives, preferably used at right angles at each successive stage until the scratches are no longer visible. Further smoothing of the surface produces a surface film known as the Beilby layer. Modem belief still accepts the presence of this Beilby layer, but it is now thought to be produced by a filling in of the scratches by the powdered particulate being removed from the surface, and is described as a microcrystalline layer (Phillips, 1982). The difference between a polishing and abrasive agent is rather obscure. Indeed, some polishing agents will remove a considerable amount of material. Further evidence of the enigma is exemplified by two authorities (Williams and Cunningham, 1979) who claim that polishing of the amalgam leaves the surface more homogeneous thereby resisting corrosion. Unfortunately, the exact process whereby such a homogeneous layer is produced is still left rather vague. Turning now to the procedure of actual polishing, what is being sought is the achievement of two states-first the production of a mirror-like surface from which all scratches have been eliminated and, secondly, the production of margins which are virtually undetectable to a dental probe. These can only be accomplished by the expenditure of time and effort. An informal survey has shown that using the customary armamentarium of stones, finishing burs, pumice and a buffing agent (such as zinc oxide), a time commitment of 34 min can be incurred, even for the simplest filling. More complex restorations may require double this time. There is thus a considerable additional cost to be justified by the individual dentist, and by the Health Service or thirdparty carrier, for the prescription of the extra stage of surface polishing. The timing of when polishing should be carried out is, to an extent, controversial. Usually it is performed when it is convenient to the operator with the proviso that at least 24 h have elapsed since insertion of the tilling. It is presumed that the peritectic reaction involving the different phases of the amalgam will have taken place by this time or, at any rate, be of such a degree that disruption of the surface by the polishing is unlikely. Yet having said that, there have been recent reports in the literature of suggestions that ‘polishing’ should be attempted almost immediately after insertion of the amalgam. The surface properties of amalgams that have been polished 8 min alter insertion have been examined and compared with surfaces polished after a conventional interval, the authors maintaining that a smoother texture was observed on the immediately polished restorations (Lyon et al., 1982; Yi et al., 1984). The benefits associated with this texture are not, however, proven.
Jeffrey and Pit-b: Finishing of amalgam restorations
Other claimed advantages of polishing have been studied and described. It has been shown that the Vickers hardness of the surface of amalgams can be improved from 75 to 90 by polishing and, in addition, the transverse strength improves (Venz, 1982). While it is commonly believed that polishing will reduce the opportunity for corrosion to occur, it has to be pointed out that should areas of unequal polishing occur, a difference in surface electrical potential becomes possible and tarnishing will then occur to a greater extent in the lesser polished portion. Another of the paradoxes associated with polishing is the contrast between the great deal of attention which has been focused on the quality of surface finish attained for the occlusal surfaces and the scant attention which has been paid to the finishing of approximal surfaces restored with amalgam. It would seem that it is quite feasible to obtain a good polish on these approximal amalgam surfaces without tooth separation (McLundie et al., 1988), although once again the absolute need to do so is unproven.
BURNISHING The technique of burnishing is often applied during descriptions of amalgam manipulation without clearly stating what is meant by this term. Burnishing an inlay, for example, generally infers the physical movement of the gold so that a margin may be bent to give closer adaptation to the tooth surface. Usually a measure of force accompanies this technique. On the other hand, bumishing is often produced by a rotary instrument, such as an engine burnisher, which uses very little applied force. In the context of amalgam usage, however, burnishing can also mean the rapid rubbing of a smooth instrument over setting, or indeed set, material to produce a smooth shiny surface. In the case of burnishing setting amalgam little force is used, cotton wool even being employed as the burnisher. Within the past lo-15 years, opinions regarding the burnishing of amalgams have changed. Burnishing used to be deprecated on two accounts-the well-known beliefs that it brought the mercury-rich layer to the surface, and the lesser known risk of burnishing in one direction being the cause of opening the opposite margin of the filling. Burnishing is now once again recommended and we feel that some further elaboration on this topic must be given. The timing of burnishing is important. Research reports by Kato et al. (1968) and by May et al. (1983) give two time sequences, the former describing the effects after 7 and 12 min following trituration, the latter giving the more practical and understandable times of precarving and postcarving. Precarved burnishing is really a continuation of the condensation process and has been shown to be more effective in reducing the mercuric content than conventional condensation (May et al., 1983). In addition,
57
it is said to improve the carving characteristics of the amalgam and the authors make the statement that, ‘It is possible that precarved burnishing may offer greater resistance to marginal failure than conventional polishing’. Postcarved burnishing will reduce surface pits; after 2 years, restorations in these studies were found to be pit free. A satin or velvet fmish should be the aim at this stage, not a highly reflective surface. Among the properties claimed to be achieved by postcarved burnishing are the following: increased surface hardness, a reduction in porosity and residual mercury, a slower rate of corrosion and an improvement in the marginal seal. One further study (Guertsen and Hartman, 1985) has shown that following the carving of amalgam, the marginal cleft remaining gave a recorded depth of 60 ,um and that burnishing with metal instruments reduced this to 20 pm.
CAVITY
PREPARATION
AND CARVING
In any consideration of the finishing of amalgam restorations, it is imperative to integrate the roles that, first, cavity preparation and, secondly, the carving of the restoration immediately after its insertion play in the final establishment of the restoration and its margins. These factors, however, are subjects in themselves and therefore we will only make reference here to factors which have a direct bearing on the actual polishing of the amalgam. The type of amalgam used in the restoration is also important insofar that the carvability and early compressive strength will influence the amount of finishing that can take place in the immediate postinsertion period. With regard to carving, it is difficult to find in the literature specific instruction about the extent to which this should be carried out. Traditionally, students have been taught, by means of demonstrations and photographs of completed restorations, that a series of deep fissures crossing the surface of the restoration duplicating, or in some cases ‘improving-on’, the anatomical form of the occlusal surface was the ideal for which they should strive. Anything short of this was judged to be (to a degree) a compromised failure. If, however, deep carved fissures are not introduced, but instead the occlusal surface of fillings is left relatively flat (while maintaining occlusal stops and ensuring that there are no occlusal interferences in all mandibular excursions), it is possible to raise the amalgam margin angle (AMA) from its customary mean of 64” to somewhere nearer the acceptable figure of 70”, or above. The figure of 70” has been recommended as the minimum that offers a reasonable chance of the amalgam remaining intact during function over long periods (Elderton, 1984). In addition, it is the opinion of Mahler (1967) that any deep carving of the occlusal surface will invite the risk of a loss of bulk in areas likely to develop high tensile stress. The detailed carving of a restoration may be thought to allow a greater penetration of the food rather than an
58
J. Dent. 1989; 17: No. 2
improvement in masticatory destruction of the food bolus per se. Unless the amalgam restoration is very extensive, however, there is usually enough cuspal form left to allow penetration of the food bolus to take place easily, so perhaps diminishing the necessity for further elaboration of the cuspal pattern in the amalgam. Pickard (1978) advocates carving definite marginal pits adjacent to the approximal marginal ridges as he believes them to be important in deflecting food from the contact areas. One is left, nevertheless, with the reluctant conclusion that carving is carried out more with an aesthetic than a functional purpose in mind. It is, however, important on at least five accounts. 1. Sufficient excess amalgam (used in obtaining complete condensation of the material within the cavity) must be removed in order to leave an amalgam that is as free of mercury as possible. 2. The marginal ridge on approximal restorations must be carved so as to integrate with the height of the adjacent tooth (failure in this area is one of the most common errors seen in students’ finishing of posterior restorations) and to ensure that an adequate contact with the neighbouring tooth is established. 3. The correct trimming of the gingival margin(s) of Class II type restorations is profoundly important, not only from the recurrent caries point of view, but also from the potential of continuing plaque-mediated loss of tooth attachment (see below). 4. Where the amalgam extends far enough up cusp inclines to involve functional occlusal areas it is important on the one hand that the height and contour of the restoration provides adequate occlusal stops while on the other ensuring that occlusal interferences are not introduced. 5. Proper axial contouring of the Class V type of restoration was held to be very important as it was felt that the wrong degree of curvature could be the precursor of periodontal breakdown (Hopwood et al., 1984), although many now feel that the patient’s plaque control must be the dominant factor in the periodontal prognosis. The extent of the problem of achieving a satisfactory contour at the gingival margin(s) of Class II restorations is illustrated in one survey carried out in Dundee (Than et al., 1982) in which 240 extracted teeth, each with a restoration on one approximal surface only, were examined. It was deemed that only 27 per cent of the margins were of good quality; 60 per cent had overhangs and 13 per cent exhibited a frank deficiency. In other words almost three-quarters of the cervical margins of the Class II restorations examined were of less than good quality in terms of the degree of accuracy achieved in placing and shaping the amalgam. There is a considerable amount of scientific evidence (Alexander, 1968; Gilmore and Sheiham, 197 1) which argues that the presence of overhanging ledges in amalgam restorations will encourage plaque formation which will often lead to gingival and periodontal inflammation.
Fig. 7. Diagrammatic illustration of an upper first premolar showing the poor adaptation of a matrix band-or dental floss-across the region of the canine fossa (shaded).
It is commonly believed that dental floss should be used to ensure the gingival margin is as clear of ledges as possible. However, if this is done at too early a stage in the procedure there is a danger of spoiling the marginal ridge, or a contact point. It must also be appreciated that the manipulation of floss will not deal with any concavity in the horizontal plane (see Fig. I) as, like the matrix band, the floss will only touch the tooth structure at the highest points of the gingival contour.
MARGINAL
ADAPTATION
Attention must now be given to the most vulnerable part of any restoration, namely the cavosurface junction. While the adaptation at the cervical margin has already been referred to, the integrity of the cavosurface junction on the occlusal surfaces and at the embrasures must also be considered. The traditional view has been that a nearperfect adaptation was attainable and should be a realistic goal. The alternative view recently restated by Matsuda and Fusayama (1970), and endorsed by many clinicians, is that all margins are defective to a greater or lesser extent-it is only a matter of degree. The difftculty comes in deciding what ‘tolerance’ should be considered a realistic optimum target to strive for and what maximum ‘threshold’ of tolerance should be set after which a decision to replace the restoration becomes mandatory? It should be appreciated that not all imperfect margins are necessarily harmful. An examination of any natural tooth will readily reveal the existence of fissures and crevices rougher and deeper than those found around many restorations and yet caries and periodontal degeneration do not inevitably occur at these sites. The likelihood for plaque retention at these sites is enhanced but in a great number of cases this can be adequately accommodated by the body’s natural defences aided in many cases by good oral hygiene procedures. Could not, therefore, the same slightly complacent but acceptable attitude apply to many of the less than perfect margins of restorations which many clhi,cians currently suspect must be harmful?
Jeffrey and Pitts: Finishing of amalgam restorations
Should any initial dimensional change, due to creep or expansion, occur at a margin, this can be corrected by polishing. This should mean that polished amalgams will exhibit superior marginal integrity. However, this view is not supported by the results of one investigation (Mayhew and Schmeltzer, 1985) which studied the differences in marginal integrity between polished and unpolished highcopper amalgams. The findings suggest that (up to an admittedly brief period of 3 years following insertion) polishing of high-copper alloys may have little effect, positively or negatively, on the marginal integrity of restorations. It is by pursuing the correction of initial marginal imperfections that the act of polishing may exert its greatest overall benefit to the restorative procedure. The procedure atfords the operator the opportunity of reviewing each part of the margin in turn and of removing any excess, or correcting any errors missed at the carving stage. A danger, on the other hand, is that further removal of amalgam at parts of the margin which are already overcarved may increase the amalgam margin angle so as to predispose that particular part of the margin to early material fracture.
POLISHING
AND SECONDARY
CARIES
Considering the frequency with which restorations are replaced following a diagnosis of secondary caries, as well as the amounts of money (at a national level) regularly expended to provide these replacement restorations, it is surprising just how little is known about the reliability and validity of the methods used to diagnose the secondary carious lesion (Kidd, 1989). A recent survey carried out on the durability of amalgam restorations by the Scandinavian Institute of Dental Materials is enlightening (Mjor, 1985). In this study 650 Class II amalgam restorations were placed by six clinicians. The completed restorations were examined immediately after polishing, after a further 6 months, and then once a year, using direct and indirect clinical techniques. A retrospective analysis of the reasons for failure was then carried out. The most significant result to emerge from this work was that of those restorations with secondary caries, 94 per cent of the carious lesions were found at the gingival margin. The cervical margin is an area that is most difficult to gain access to and thus to polish. This finding may well be more related to the problems of the relatively isolated and protected environment in which this part of the margin is situated rather than the surface polish of the amalgam in this area per se. The anatomy of the interproximal space favours aggregations of mature cariogenic plaque. This, coupled with the marginal imperfections associated with the difficulty of finishing this part of cavity preparations, as well as difficulties in successfully adapting a matrix and properly condensing amalgam in this area, increases the risk of secondary caries at cervical sites.
SUMMARY
59
AND DISCUSSION
It is difficult to reconcile the traditional views regarding the need to polish amalgam restorations to a mirror-like surface with the degree of finishing often performed outside dental schools. However, the objective need for such a surface fmish, in terms of compromised durability of restorations which are satisfactory in other respects, has not been demonstrated. Similarly the need for, and even the desirability of, a highly carved occlusal surface with a well-demarcated fissure pattern is increasingly being questioned and compared to the need to maintain high amalgam margin angles. The choice of amalgam type is most important, not only from the convenience of a working time that allows an adequate opportunity to complete carving and burnishing, but by the increased edge strength offered by the highcopper alloys. This choice may indeed have a greater overall bearing on longevity than polishing. Burnishing is now again accepted as being advisable and should be carried out, not as a final gesture that the filling is complete, but with a definite and positive intent. While many operators zealously seek out any evidence of marginal defects, the clinical significance of any small gap between amalgam and tooth structure (particularly on the occlusal surface) is equivocal. The long-term prognosis associated with various sizes of such disunions, if they are monitored and/or treated in a preventive way, are not well understood at present. Although secondary caries is often cited as a reason to replace a restoration which has a defective occlusal margin, recent evidence challenges the validity of this argument. The researchers in the study by Mjor (1985) made the claim that no direct relationship appears to exist between occlusal marginal degradation and secondary caries. If this is true-and it is worth noting that some of their occlusal margins did become less than perfect in the passage of time but did not acquire secondary cariesthen the expenditure of costly time in polishing margins at this particular site can be seriously questioned. Indeed many unpolished occlusal margins do seem to survive for many years without the benefit of a mirrorlike sheen. Interestingly, a study on marginal fracture being a predictor to the longevity of two types of dental amalgams, carried out over a lo-year period, found that although there was a significant difference in the marginal deterioration between the different alloys, there was no significant difference in respect of the amount of recurrent caries that ensued (Hamilton et al., 1983). It must, however, be admitted that this finding was to a degree offset by the work of Goldberg et al. (1981). It now seems that the majority of secondary carious lesions actually affect the cervical margin of amalgam restorations. Greater attention to margin quality in this area would seem to be extremely important. However, this issue is complicated by the difficulty in reliably diagnosing secondary caries at various sites.
60
J. Dent. 1989;
17: No. 2
It would seem that a great many amalgam restorations are replaced because, after a comparatively short period, they fail to match up to the traditional ‘ideal’ or ‘textbook concepts which are used (sometimes unconsciously) as a reference standard by many operators. A very great deal of tooth structure, time and money is expended on replacing restorations but surprisingly little research has been carried out attempting to define and test criteria to establish exactly when restorations should be replaced. There is then a need for further research in this difficult, rather neglected, but important area in order to establish reliable criteria for the diagnosis and assessment of marginal gaps as well as the presence of (and degree of activity of) secondary caries. Such studies will hopefully lead to a more scientific basis for deciding upon the optimal standard of finishing for the initial placement of an amalgam restoration. They would also help to ascertain what level of imperfection is acceptable when deciding whether or not to replace an existing restoration. In the interim it would seem advisable for both teachers and practitioners to strive for a balance between the high technical standards which should be used in the initial placement of restorations and the more, circumspect limited intervention approach to making treatment decisions for restorations which, although short of the traditional ideal, can still provide functional service. Acknowledgements The authors are indebted to the photographic section of the University’s Central Media Service for Fig. 1. Dr Pitts acknowledges financial support from the Scottish Home and Health Department’s Chief Scientist Office.
References Alexander A. G. (1968) Periodontal aspects of conservative dentistry. Br. Dent. J. 125, 111-114. Elderton R J. (1984) Cavo-surface angles, amalgam margin angles and occlusal cavity preparations. Br. Dent. J. 156,3 19-324. Geurtsen W. K. H. and Hartman U. (1985) Marginal adaptation of amalgam restorations after two different finishing methods. J. Dent. Res. 64, 241. Gilmore N. and Sheiharn A. (197 1) Overhanging restorations and periodontal disease. J. Periodontol. 42, 8-12.
Goldberg J., Tamer J., Munster E. et al. (1981) Crosssectional clinical evaluation of recurrent enamel caries, restoration of marginal integrity and oral hygiene status. J. Am. Dent. Assoc. 102, 635. Hamilton J. C., Moffa J. G., Ellison J. A. et al. (1983) Marginal fracture not a predictor of longevity for two dental amalgam alloys-a 10 year study. J. Prosthet. Dent. 50, 200-202. Hopwood W. A., Lilley J. D. and Nixon G. S. (1984) Gingival tissue response to the position and contour of metal-ceramic crown margins. J. Dent. Res. 63, 5 10. Kato S., Okuse K. and Fusayama T. (1968) The effect of burnishing on the marginal seal of amalgam restorations. J, Prosthet. Dent. 19, 393-398. Kidd E. A. M. (1989) Caries diagnosis within restored teeth. In: Anusavice K. J. (ed.), Quality Evaluation of Dental Restorations: Criteria for Placement and Replacement. Chicago, Quintessence (in press). Lyon H. E., Hawks M. L. and Mitchell R J. (1985) The roughness of high copper amalgams after immediate polishing. J. Dent. Res. 64, 180. McLundie A. C., Patterson C. J. W. and Stirrups D. R (1988) Approximal surface finishes on amalgam restorations. Restorative Dent. 4, 18-22. Mahler D. B. (1967) Physical properties and manipulation of amalgam. Dent. Clin. North Am. 11, 213. Matsuda N. and Fusayama T. (1970) Marginal failure of amalgam restorations. J. Prosthet. Dent. 23, 658-661. May K. N., Wilder A. D. and Leinfelder K. F. (1983) Burnished amalgam restorations-a 2 year clinical evaluation. J. Prosthet. Dent. 49, 193-197. Mayhew R B. and Scluneltzer C. D. (1985) Evaluation of polishing techniques in marginal integrity of amalgam restorations. J. Dent. Res. 64, 278. Mjor I. A. (1985) Clinical assessment of amalgam restorations. The Dental Materials Coderence, St Andrews, Scotland. 19-20 September 1985. Dundee, University of Dundee, pp. 8-9. Phillips R W. (ed.) (1982) Skinner’s Source of Dental Materials, 8th edn. Philadelphia, W. B. Saunders, p. 586. Pickard H. M. (1978) A Manual of Operative Dentistry, 6th edn. Oxford, Oxford University Press, p. 86. Than A., Duguid R and McKendrick A. J. W. (1982) Relationship between restorations and the level of the periodontal attachment. J. Clin. Periodontol. 9, 193-202. Venz S. (1982) Mechanical strength and Polishing behaviour of dental amalgams. J. Dent. Res. 61, 269. Williams D. F. and Cunningham J. (1979) Materials in Clinical Dentistry. Oxford, Oxford University Press, p. 138. Yi J., Dennison J. B. and Asgar K. (1985) The effect of early polishing procedures on a variety of amalgam alloys. J. Dent. Res. 63, 117.
Correspondence should be addressed to: Dr I. W. M. Jeffrey, Department Dundee DDl 4HR UK.
of Conservative
Dentistry, Dental School, Park Place,
1989;
55
17: 55-60
Finishing of amalgam restorations: what degree is it necessary?
to
I. W. M. Jeffrey and N. B. Pit&* Department of Conservative Dentistry and * Dental Health Services Research Unit Department of Periodontology and Community Dentistry, Dental School, Dundee, UK KEY WORDS: Amalgam, Operative technique, Review
J. Dent. 1989;
17: 55-60
(Received 26 May 1988;
reviewed 7 July 1988;
accepted 18 July 1988)
ABSTRACT This paper attempts to elicit from the literature evidence to guide the clinician in deciding what degree of finishing and polishing is appropriate for amalgam restorations. Much of the justification for the polishing of amalgams has, in the past, been empirically based. Comprehensive scientific investigation into the validity of this practice does, however, present a number ofmethodological problems which have yet to be successfully overcome. An attempt is made to define what is meant by the term ‘polished’ (in a clinical context) while the benefits of polishing, as far as they are known, are discussed. The roles that cavity preparation, carving and burnishing play in the finishing of a restoration are reviewed, as is the conflicting evidence regarding their respective advantages and disadvantages. The importance of marginal adaptation, the relationship of polishing to secondary caries, and the dilemma in choosing (in appropriate cases) between the replacement of a restoration or its improvement by polishing are also considered.
INTRODUCTION
In operative
dentistry it is commonly regarded as a sine qua non that amalgam restorations, at some time following their placement, should be polished. The justification for the adoption of such a polishing regimen has been based largely on pragmatic beliefs and a priori
arguments, the most obvious of these being the propensity for unpolished aspects of the restoration to encourage plaque formation. This outcome, it is assumed, will lead to the subsequent degradation of the cavity margins by secondary caries which will necessitate replacement of the restoration. In the undergraduate context, and with apparently little variation between the dental schools, there is a place on the student’s record card of work carried out wherein confirmation is given that each restoration inserted has been duly polished. This ‘polishing’ is often specifically graded by supervizing staff. In general practice, however, the state of finish of restorations would appear to be a tripartite matter between the individual beliefs and conscience of the practitioner; the occasional introverted curiosity of the more dentally aware patient; and the surveillance of ‘third-party assessors’ such as a Dental Reference Officer in the UK. In both dental schools and 0 1989 Butterworth 8c Co. Publishers 0300-5712/89/020055-06 $03.00
Ltd.
general practice, however, many dentists apparently regard the absence of a highly polished amalgam surface (elaborately carved to resemble the majority of the occlusal anatomical pits and fissures), or the presence of any slight marginal defect, as ample reason to initiate replacement of the restoration. It must surely be the responsibility of academic dentistry to investigate the fundamental, but routine problem of identifying both the optimal and the acceptable level to which an amalgam restoration should be finished. Investigative studies assessing the need for finishing, determining the effects of (and costs associated with) achieving varying standards of surface polishing and marginal integrity, and ascertaining the sequelae to be expected if certain standards are not met in a representative clinical environment, are all highly desirable. There are, however, significant methodological problems which must be overcome first. The major difficulty concerns the large number of variables to be considered in any such study-the oral hygiene and caries risk of the hosts, the siting and dimension of the particular restorations, the choice and standardization of the finishing techniques employed, the types of material used, the level of skill and representativeness of the operator(s), to name but a few.
56
J. Dent. 1989; 17: No. 2
Very large numbers of patient subjects would have to be employed before proper statistical control could vouchsafe a correct conclusion being made. Another complication is the well-established observation that a great many restorations which are not polished or carved to any great extent exist for long periods without evidence of overt manifestations of harm. The aim of this paper is to examine current opinion with respect to some of the various stages of finishing amalgam restorations and to discuss the appearance of amalgam restorations as a factor influencing the treatment decision to replace them.
POLISHING It is commonly believed to be advisable to polish any material that is to be inserted ‘permanently’ in the mouth because surface roughness predisposes to plaque accumulation and an increased risk of caries; however, polishing has not always been carried out. A simple example is seen in the finishing of the fitting surfaces of metal partial dentures with surfaces contiguous to tooth tissue. No more than a vigorous cleaning and a removal of gross roughnesses from the casting is performed, but in no sense can this be considered true polishing. It is conceded that in this case the denture is removable but it is, nevertheless, eligible to remain in the oral environment for long periods in direct apposition to surfaces susceptible to caries. In addition, the polishing of any material per se in the mouth has been motivated by other factors. These include the presence of sharp, rough or jagged surfaces which could be unpleasant for the patient and which could initiate either acute or chronic damage to the oral mucosa; or food trapping which in turn can alter the surface of metallic restorative materials leading to electrolytically sponsored corrosion and staining. However, the clinical significance of these problems, particularly with respect to the modern high-copper amalgam alloys, in terms of their effect upon the longevity of restorations, is not well understood. One difficulty that arises in any discourse on polishing within the mouth is to define, in exact terms, what is meant by polishing. What, for example, distinguishes a smooth surface from a polished one? It is apparent that, at a technical level, definitive differences are rather subtle and slightly confusing. There is almost a temptation to deal in simplistic terms, such as rough or smooth, until it is realized that these terms are even more confusing, arbitrary and ambiguous. The process of polishing should produce a shiny, mirror-like (when viewed by the naked eye) surface without the production or assistance of supplemental surface lihns created in other situations by the addition of wax or lacquer. In essence, polishing means the formation of a scratch-free surface; the scratches varying from the deep macroscopically evident scores, to fine lines which require a microscope to see.
Only on the final disappearance of all scratches can the surface be said to be truly polished. Even today, the polishing process does not seem to be clearly understood. The usually accepted sequence of events believes that the initial stage is one of abrasion, where the peaks of the score furrows are reduced in height by the action of a series of increasingly fmer abrasives, preferably used at right angles at each successive stage until the scratches are no longer visible. Further smoothing of the surface produces a surface film known as the Beilby layer. Modem belief still accepts the presence of this Beilby layer, but it is now thought to be produced by a filling in of the scratches by the powdered particulate being removed from the surface, and is described as a microcrystalline layer (Phillips, 1982). The difference between a polishing and abrasive agent is rather obscure. Indeed, some polishing agents will remove a considerable amount of material. Further evidence of the enigma is exemplified by two authorities (Williams and Cunningham, 1979) who claim that polishing of the amalgam leaves the surface more homogeneous thereby resisting corrosion. Unfortunately, the exact process whereby such a homogeneous layer is produced is still left rather vague. Turning now to the procedure of actual polishing, what is being sought is the achievement of two states-first the production of a mirror-like surface from which all scratches have been eliminated and, secondly, the production of margins which are virtually undetectable to a dental probe. These can only be accomplished by the expenditure of time and effort. An informal survey has shown that using the customary armamentarium of stones, finishing burs, pumice and a buffing agent (such as zinc oxide), a time commitment of 34 min can be incurred, even for the simplest filling. More complex restorations may require double this time. There is thus a considerable additional cost to be justified by the individual dentist, and by the Health Service or thirdparty carrier, for the prescription of the extra stage of surface polishing. The timing of when polishing should be carried out is, to an extent, controversial. Usually it is performed when it is convenient to the operator with the proviso that at least 24 h have elapsed since insertion of the tilling. It is presumed that the peritectic reaction involving the different phases of the amalgam will have taken place by this time or, at any rate, be of such a degree that disruption of the surface by the polishing is unlikely. Yet having said that, there have been recent reports in the literature of suggestions that ‘polishing’ should be attempted almost immediately after insertion of the amalgam. The surface properties of amalgams that have been polished 8 min alter insertion have been examined and compared with surfaces polished after a conventional interval, the authors maintaining that a smoother texture was observed on the immediately polished restorations (Lyon et al., 1982; Yi et al., 1984). The benefits associated with this texture are not, however, proven.
Jeffrey and Pit-b: Finishing of amalgam restorations
Other claimed advantages of polishing have been studied and described. It has been shown that the Vickers hardness of the surface of amalgams can be improved from 75 to 90 by polishing and, in addition, the transverse strength improves (Venz, 1982). While it is commonly believed that polishing will reduce the opportunity for corrosion to occur, it has to be pointed out that should areas of unequal polishing occur, a difference in surface electrical potential becomes possible and tarnishing will then occur to a greater extent in the lesser polished portion. Another of the paradoxes associated with polishing is the contrast between the great deal of attention which has been focused on the quality of surface finish attained for the occlusal surfaces and the scant attention which has been paid to the finishing of approximal surfaces restored with amalgam. It would seem that it is quite feasible to obtain a good polish on these approximal amalgam surfaces without tooth separation (McLundie et al., 1988), although once again the absolute need to do so is unproven.
BURNISHING The technique of burnishing is often applied during descriptions of amalgam manipulation without clearly stating what is meant by this term. Burnishing an inlay, for example, generally infers the physical movement of the gold so that a margin may be bent to give closer adaptation to the tooth surface. Usually a measure of force accompanies this technique. On the other hand, bumishing is often produced by a rotary instrument, such as an engine burnisher, which uses very little applied force. In the context of amalgam usage, however, burnishing can also mean the rapid rubbing of a smooth instrument over setting, or indeed set, material to produce a smooth shiny surface. In the case of burnishing setting amalgam little force is used, cotton wool even being employed as the burnisher. Within the past lo-15 years, opinions regarding the burnishing of amalgams have changed. Burnishing used to be deprecated on two accounts-the well-known beliefs that it brought the mercury-rich layer to the surface, and the lesser known risk of burnishing in one direction being the cause of opening the opposite margin of the filling. Burnishing is now once again recommended and we feel that some further elaboration on this topic must be given. The timing of burnishing is important. Research reports by Kato et al. (1968) and by May et al. (1983) give two time sequences, the former describing the effects after 7 and 12 min following trituration, the latter giving the more practical and understandable times of precarving and postcarving. Precarved burnishing is really a continuation of the condensation process and has been shown to be more effective in reducing the mercuric content than conventional condensation (May et al., 1983). In addition,
57
it is said to improve the carving characteristics of the amalgam and the authors make the statement that, ‘It is possible that precarved burnishing may offer greater resistance to marginal failure than conventional polishing’. Postcarved burnishing will reduce surface pits; after 2 years, restorations in these studies were found to be pit free. A satin or velvet fmish should be the aim at this stage, not a highly reflective surface. Among the properties claimed to be achieved by postcarved burnishing are the following: increased surface hardness, a reduction in porosity and residual mercury, a slower rate of corrosion and an improvement in the marginal seal. One further study (Guertsen and Hartman, 1985) has shown that following the carving of amalgam, the marginal cleft remaining gave a recorded depth of 60 ,um and that burnishing with metal instruments reduced this to 20 pm.
CAVITY
PREPARATION
AND CARVING
In any consideration of the finishing of amalgam restorations, it is imperative to integrate the roles that, first, cavity preparation and, secondly, the carving of the restoration immediately after its insertion play in the final establishment of the restoration and its margins. These factors, however, are subjects in themselves and therefore we will only make reference here to factors which have a direct bearing on the actual polishing of the amalgam. The type of amalgam used in the restoration is also important insofar that the carvability and early compressive strength will influence the amount of finishing that can take place in the immediate postinsertion period. With regard to carving, it is difficult to find in the literature specific instruction about the extent to which this should be carried out. Traditionally, students have been taught, by means of demonstrations and photographs of completed restorations, that a series of deep fissures crossing the surface of the restoration duplicating, or in some cases ‘improving-on’, the anatomical form of the occlusal surface was the ideal for which they should strive. Anything short of this was judged to be (to a degree) a compromised failure. If, however, deep carved fissures are not introduced, but instead the occlusal surface of fillings is left relatively flat (while maintaining occlusal stops and ensuring that there are no occlusal interferences in all mandibular excursions), it is possible to raise the amalgam margin angle (AMA) from its customary mean of 64” to somewhere nearer the acceptable figure of 70”, or above. The figure of 70” has been recommended as the minimum that offers a reasonable chance of the amalgam remaining intact during function over long periods (Elderton, 1984). In addition, it is the opinion of Mahler (1967) that any deep carving of the occlusal surface will invite the risk of a loss of bulk in areas likely to develop high tensile stress. The detailed carving of a restoration may be thought to allow a greater penetration of the food rather than an
58
J. Dent. 1989; 17: No. 2
improvement in masticatory destruction of the food bolus per se. Unless the amalgam restoration is very extensive, however, there is usually enough cuspal form left to allow penetration of the food bolus to take place easily, so perhaps diminishing the necessity for further elaboration of the cuspal pattern in the amalgam. Pickard (1978) advocates carving definite marginal pits adjacent to the approximal marginal ridges as he believes them to be important in deflecting food from the contact areas. One is left, nevertheless, with the reluctant conclusion that carving is carried out more with an aesthetic than a functional purpose in mind. It is, however, important on at least five accounts. 1. Sufficient excess amalgam (used in obtaining complete condensation of the material within the cavity) must be removed in order to leave an amalgam that is as free of mercury as possible. 2. The marginal ridge on approximal restorations must be carved so as to integrate with the height of the adjacent tooth (failure in this area is one of the most common errors seen in students’ finishing of posterior restorations) and to ensure that an adequate contact with the neighbouring tooth is established. 3. The correct trimming of the gingival margin(s) of Class II type restorations is profoundly important, not only from the recurrent caries point of view, but also from the potential of continuing plaque-mediated loss of tooth attachment (see below). 4. Where the amalgam extends far enough up cusp inclines to involve functional occlusal areas it is important on the one hand that the height and contour of the restoration provides adequate occlusal stops while on the other ensuring that occlusal interferences are not introduced. 5. Proper axial contouring of the Class V type of restoration was held to be very important as it was felt that the wrong degree of curvature could be the precursor of periodontal breakdown (Hopwood et al., 1984), although many now feel that the patient’s plaque control must be the dominant factor in the periodontal prognosis. The extent of the problem of achieving a satisfactory contour at the gingival margin(s) of Class II restorations is illustrated in one survey carried out in Dundee (Than et al., 1982) in which 240 extracted teeth, each with a restoration on one approximal surface only, were examined. It was deemed that only 27 per cent of the margins were of good quality; 60 per cent had overhangs and 13 per cent exhibited a frank deficiency. In other words almost three-quarters of the cervical margins of the Class II restorations examined were of less than good quality in terms of the degree of accuracy achieved in placing and shaping the amalgam. There is a considerable amount of scientific evidence (Alexander, 1968; Gilmore and Sheiham, 197 1) which argues that the presence of overhanging ledges in amalgam restorations will encourage plaque formation which will often lead to gingival and periodontal inflammation.
Fig. 7. Diagrammatic illustration of an upper first premolar showing the poor adaptation of a matrix band-or dental floss-across the region of the canine fossa (shaded).
It is commonly believed that dental floss should be used to ensure the gingival margin is as clear of ledges as possible. However, if this is done at too early a stage in the procedure there is a danger of spoiling the marginal ridge, or a contact point. It must also be appreciated that the manipulation of floss will not deal with any concavity in the horizontal plane (see Fig. I) as, like the matrix band, the floss will only touch the tooth structure at the highest points of the gingival contour.
MARGINAL
ADAPTATION
Attention must now be given to the most vulnerable part of any restoration, namely the cavosurface junction. While the adaptation at the cervical margin has already been referred to, the integrity of the cavosurface junction on the occlusal surfaces and at the embrasures must also be considered. The traditional view has been that a nearperfect adaptation was attainable and should be a realistic goal. The alternative view recently restated by Matsuda and Fusayama (1970), and endorsed by many clinicians, is that all margins are defective to a greater or lesser extent-it is only a matter of degree. The difftculty comes in deciding what ‘tolerance’ should be considered a realistic optimum target to strive for and what maximum ‘threshold’ of tolerance should be set after which a decision to replace the restoration becomes mandatory? It should be appreciated that not all imperfect margins are necessarily harmful. An examination of any natural tooth will readily reveal the existence of fissures and crevices rougher and deeper than those found around many restorations and yet caries and periodontal degeneration do not inevitably occur at these sites. The likelihood for plaque retention at these sites is enhanced but in a great number of cases this can be adequately accommodated by the body’s natural defences aided in many cases by good oral hygiene procedures. Could not, therefore, the same slightly complacent but acceptable attitude apply to many of the less than perfect margins of restorations which many clhi,cians currently suspect must be harmful?
Jeffrey and Pitts: Finishing of amalgam restorations
Should any initial dimensional change, due to creep or expansion, occur at a margin, this can be corrected by polishing. This should mean that polished amalgams will exhibit superior marginal integrity. However, this view is not supported by the results of one investigation (Mayhew and Schmeltzer, 1985) which studied the differences in marginal integrity between polished and unpolished highcopper amalgams. The findings suggest that (up to an admittedly brief period of 3 years following insertion) polishing of high-copper alloys may have little effect, positively or negatively, on the marginal integrity of restorations. It is by pursuing the correction of initial marginal imperfections that the act of polishing may exert its greatest overall benefit to the restorative procedure. The procedure atfords the operator the opportunity of reviewing each part of the margin in turn and of removing any excess, or correcting any errors missed at the carving stage. A danger, on the other hand, is that further removal of amalgam at parts of the margin which are already overcarved may increase the amalgam margin angle so as to predispose that particular part of the margin to early material fracture.
POLISHING
AND SECONDARY
CARIES
Considering the frequency with which restorations are replaced following a diagnosis of secondary caries, as well as the amounts of money (at a national level) regularly expended to provide these replacement restorations, it is surprising just how little is known about the reliability and validity of the methods used to diagnose the secondary carious lesion (Kidd, 1989). A recent survey carried out on the durability of amalgam restorations by the Scandinavian Institute of Dental Materials is enlightening (Mjor, 1985). In this study 650 Class II amalgam restorations were placed by six clinicians. The completed restorations were examined immediately after polishing, after a further 6 months, and then once a year, using direct and indirect clinical techniques. A retrospective analysis of the reasons for failure was then carried out. The most significant result to emerge from this work was that of those restorations with secondary caries, 94 per cent of the carious lesions were found at the gingival margin. The cervical margin is an area that is most difficult to gain access to and thus to polish. This finding may well be more related to the problems of the relatively isolated and protected environment in which this part of the margin is situated rather than the surface polish of the amalgam in this area per se. The anatomy of the interproximal space favours aggregations of mature cariogenic plaque. This, coupled with the marginal imperfections associated with the difficulty of finishing this part of cavity preparations, as well as difficulties in successfully adapting a matrix and properly condensing amalgam in this area, increases the risk of secondary caries at cervical sites.
SUMMARY
59
AND DISCUSSION
It is difficult to reconcile the traditional views regarding the need to polish amalgam restorations to a mirror-like surface with the degree of finishing often performed outside dental schools. However, the objective need for such a surface fmish, in terms of compromised durability of restorations which are satisfactory in other respects, has not been demonstrated. Similarly the need for, and even the desirability of, a highly carved occlusal surface with a well-demarcated fissure pattern is increasingly being questioned and compared to the need to maintain high amalgam margin angles. The choice of amalgam type is most important, not only from the convenience of a working time that allows an adequate opportunity to complete carving and burnishing, but by the increased edge strength offered by the highcopper alloys. This choice may indeed have a greater overall bearing on longevity than polishing. Burnishing is now again accepted as being advisable and should be carried out, not as a final gesture that the filling is complete, but with a definite and positive intent. While many operators zealously seek out any evidence of marginal defects, the clinical significance of any small gap between amalgam and tooth structure (particularly on the occlusal surface) is equivocal. The long-term prognosis associated with various sizes of such disunions, if they are monitored and/or treated in a preventive way, are not well understood at present. Although secondary caries is often cited as a reason to replace a restoration which has a defective occlusal margin, recent evidence challenges the validity of this argument. The researchers in the study by Mjor (1985) made the claim that no direct relationship appears to exist between occlusal marginal degradation and secondary caries. If this is true-and it is worth noting that some of their occlusal margins did become less than perfect in the passage of time but did not acquire secondary cariesthen the expenditure of costly time in polishing margins at this particular site can be seriously questioned. Indeed many unpolished occlusal margins do seem to survive for many years without the benefit of a mirrorlike sheen. Interestingly, a study on marginal fracture being a predictor to the longevity of two types of dental amalgams, carried out over a lo-year period, found that although there was a significant difference in the marginal deterioration between the different alloys, there was no significant difference in respect of the amount of recurrent caries that ensued (Hamilton et al., 1983). It must, however, be admitted that this finding was to a degree offset by the work of Goldberg et al. (1981). It now seems that the majority of secondary carious lesions actually affect the cervical margin of amalgam restorations. Greater attention to margin quality in this area would seem to be extremely important. However, this issue is complicated by the difficulty in reliably diagnosing secondary caries at various sites.
60
J. Dent. 1989;
17: No. 2
It would seem that a great many amalgam restorations are replaced because, after a comparatively short period, they fail to match up to the traditional ‘ideal’ or ‘textbook concepts which are used (sometimes unconsciously) as a reference standard by many operators. A very great deal of tooth structure, time and money is expended on replacing restorations but surprisingly little research has been carried out attempting to define and test criteria to establish exactly when restorations should be replaced. There is then a need for further research in this difficult, rather neglected, but important area in order to establish reliable criteria for the diagnosis and assessment of marginal gaps as well as the presence of (and degree of activity of) secondary caries. Such studies will hopefully lead to a more scientific basis for deciding upon the optimal standard of finishing for the initial placement of an amalgam restoration. They would also help to ascertain what level of imperfection is acceptable when deciding whether or not to replace an existing restoration. In the interim it would seem advisable for both teachers and practitioners to strive for a balance between the high technical standards which should be used in the initial placement of restorations and the more, circumspect limited intervention approach to making treatment decisions for restorations which, although short of the traditional ideal, can still provide functional service. Acknowledgements The authors are indebted to the photographic section of the University’s Central Media Service for Fig. 1. Dr Pitts acknowledges financial support from the Scottish Home and Health Department’s Chief Scientist Office.
References Alexander A. G. (1968) Periodontal aspects of conservative dentistry. Br. Dent. J. 125, 111-114. Elderton R J. (1984) Cavo-surface angles, amalgam margin angles and occlusal cavity preparations. Br. Dent. J. 156,3 19-324. Geurtsen W. K. H. and Hartman U. (1985) Marginal adaptation of amalgam restorations after two different finishing methods. J. Dent. Res. 64, 241. Gilmore N. and Sheiharn A. (197 1) Overhanging restorations and periodontal disease. J. Periodontol. 42, 8-12.
Goldberg J., Tamer J., Munster E. et al. (1981) Crosssectional clinical evaluation of recurrent enamel caries, restoration of marginal integrity and oral hygiene status. J. Am. Dent. Assoc. 102, 635. Hamilton J. C., Moffa J. G., Ellison J. A. et al. (1983) Marginal fracture not a predictor of longevity for two dental amalgam alloys-a 10 year study. J. Prosthet. Dent. 50, 200-202. Hopwood W. A., Lilley J. D. and Nixon G. S. (1984) Gingival tissue response to the position and contour of metal-ceramic crown margins. J. Dent. Res. 63, 5 10. Kato S., Okuse K. and Fusayama T. (1968) The effect of burnishing on the marginal seal of amalgam restorations. J, Prosthet. Dent. 19, 393-398. Kidd E. A. M. (1989) Caries diagnosis within restored teeth. In: Anusavice K. J. (ed.), Quality Evaluation of Dental Restorations: Criteria for Placement and Replacement. Chicago, Quintessence (in press). Lyon H. E., Hawks M. L. and Mitchell R J. (1985) The roughness of high copper amalgams after immediate polishing. J. Dent. Res. 64, 180. McLundie A. C., Patterson C. J. W. and Stirrups D. R (1988) Approximal surface finishes on amalgam restorations. Restorative Dent. 4, 18-22. Mahler D. B. (1967) Physical properties and manipulation of amalgam. Dent. Clin. North Am. 11, 213. Matsuda N. and Fusayama T. (1970) Marginal failure of amalgam restorations. J. Prosthet. Dent. 23, 658-661. May K. N., Wilder A. D. and Leinfelder K. F. (1983) Burnished amalgam restorations-a 2 year clinical evaluation. J. Prosthet. Dent. 49, 193-197. Mayhew R B. and Scluneltzer C. D. (1985) Evaluation of polishing techniques in marginal integrity of amalgam restorations. J. Dent. Res. 64, 278. Mjor I. A. (1985) Clinical assessment of amalgam restorations. The Dental Materials Coderence, St Andrews, Scotland. 19-20 September 1985. Dundee, University of Dundee, pp. 8-9. Phillips R W. (ed.) (1982) Skinner’s Source of Dental Materials, 8th edn. Philadelphia, W. B. Saunders, p. 586. Pickard H. M. (1978) A Manual of Operative Dentistry, 6th edn. Oxford, Oxford University Press, p. 86. Than A., Duguid R and McKendrick A. J. W. (1982) Relationship between restorations and the level of the periodontal attachment. J. Clin. Periodontol. 9, 193-202. Venz S. (1982) Mechanical strength and Polishing behaviour of dental amalgams. J. Dent. Res. 61, 269. Williams D. F. and Cunningham J. (1979) Materials in Clinical Dentistry. Oxford, Oxford University Press, p. 138. Yi J., Dennison J. B. and Asgar K. (1985) The effect of early polishing procedures on a variety of amalgam alloys. J. Dent. Res. 63, 117.
Correspondence should be addressed to: Dr I. W. M. Jeffrey, Department Dundee DDl 4HR UK.
of Conservative
Dentistry, Dental School, Park Place,