0099-2399/90/1601-0009/$02.00/0 JOURNAL OF ENDODONTICS Copyright 9 1990 by The Arnencan Assooat0on of Endodontists
Printed in U.S.A.
VOL. 16, NO 1,JANUARY1990
In Vitro Quantification of the Apical Sealing Ability of Retrograde Amalgam Fillings Manabu Yoshimura, DDS, F. James Marshall, DMD, MS, and James S. Tinkle, DMD
Retrograde amalgam fillings were placed in 12 extracted human canines. The microleakage from the both directions (coronally and apically) was determined using pressurized fluid filtration at 90 rain, 6 h, 1 day, 2 days, and 1, 2, 4, and 8 wk after amalgam filling. Leakage from either direction decreased markedly in the 90-min to 6-h interval after filling. At intervals between 1 day and 8 wk, only small changes in leakage were noted. After any time period, leakage from the coronal direction was greater than leakage from the apical direction. However, this difference was not statistically significant except at 90 min when the coronal leakage was significantly greater. This experimental system was shown to reliably measure microleakage in retrograde amalgam fillings.
Retrograde amalgam seals are frequently placed when orthograde root canal fillings Jhil to reach the apical foramen or when orthograde treatments fail to heal. Surgery as an extension of endodontic therapy has been practiced ever since apicoectomy was introduced by Farrar (1). Much has been written concerning the indications and contraindications, the success rates and the risks, but little has been written about the sealing ability of apical amalgam fillings over time. The sealing ability of apical fillings has mostly been studied using freshly placed amalgam plugs (2 6). Although amalgam is thought to seal best in coronal restorations only after corrosion products are formed some months after placing (7, 8), this long-term effect has not been studied for apical amalgams. The purpose of this study was to determine if there was any change in the sealing ability of retrograde amalgam fillings related to time, using a quantitative method (9, 10).
M A ' I E R I A L S AND M E T H O D S Twelve caries-free human extracted canines were used in this study. These teeth were extracted because of periodontal
disease and were stored in water until used. Soft tissue and calculus on these teeth were removed mechanically, and the crowns were cut off at the eementoenamel junction. The pulps were removed and root canal fillings were placed in a standard fashion followed by retrograde amalgam fillings, as follows. A crown-down pressureless root canal preparation technique (11, 12) was used for cleaning and shaping the root canals to a #40 apical canal size. The true working length was 1 mm short of the radiographic apex. l-he canals were obturated with #40 gutta-percha master points, a Grossman type scaler, and accessory gutta-percha points using a lateral condensation technique. After obturation the teeth were stored in 100% relative humidity until prepared for testing. After storagc the root tips were cut off 2 mm from the apices, measured labially. Each tooth was then beveled labially at 45 degrees to the tong axis o f the root; l-mm deep cavities were prepared for amalgam fillings using a #700 tapered fiat ended fissure bur. Amalgam (Valiant PhD; L. D. Caulk, Milford, DE) was mixed according to the manufacturers instructions, placed in the prepared cavities, condensed, and burnished. The teeth were returned to 100% relative humidity for 30 min to allow amalgam initial set. ]'he gutta-percha and scaling cement were then completely removed from each canal using chloroform and files. All procedures were performed by one operator (M. Y.). The specimens were mounted, in each of two different ways, in a device (Fig. 1) designed to measure microleakage (9, 10). Microleakage was measured for each specimen with the flow pressure first applied from a coronal direction (A). The teeth were then remounted, the pressure was applied from an apical direction (B), and microleakage was remeasured. Microleakage was determined for 5-min periods, and each test was repeated three times at 90 min, 6 h, I day, 2 days, and 1, 2, 4, and 8 wk after placing the amalgam filling. Specimens were stored in distilled water between test periods and rinsed in distilled water every other day. The specimens and amalgam seals were not interfered with in any way during these many tests. Analysis of variance and Duncan's multiple test were used for statistical analysis of the validity o f the results. Selected specimens were also observed after 8 wk for the integrity of the amalgam-tooth interface using a scanning electron microscopic examination of replicas at •
Yoshimura et al.
Journal of Endodontics
FIG 1. Schematic drawings of system used to quantitate microleakage. Microleakage was determined as an amount of fluid filtration calculated from the distance that an air bubble moved in micropipette. A and B denote flow direction for two separate mountings. TABLE 1. The mean and SE of coronal and apical microleakage (#1/5 min at 10 psi) for 12 specimens* Coronal Leakage 90 min 6h 1 day 1 wk 2 wk 4 wk 8 wk
~1.13 rl[II~0.48 i;', 0.47 [ 1~\0.42 tt~0 .52 1\0.56 \0.57
• • • • • • •
0.27 - - - 0.18 0.13 0.11 0.11 0.16 0.16
Apical Leakage 0.72 • 0.14" I 0.47 • 0.12 I 0.37 • 0.08 0.34 • 0.09 ( 0.27• I 0.33 • 0.15 0.20 +_ 0.09 J
9 Repeated ttlree times at each measurement t*rne Groups combined with bars show stahst]ca/ly s~jnlf~_.ant cl,fference (p < 005).
FIG 2. a and b, scanning electron micrographs (x500) of retrograde amalgam (4) and dentin (D). Black bar shows 10 /~m. Note also interface less than 20 ~m and lack of tubules in dentin.
RESULTS DISCUSSION Table 1 shows the mean leakage of all specimens at each time studied. Coronal leakage (leakage from inside of the canal to the outside) decreased markedly after lhe 90-min recording. Leakage measured at 6 h and up to 8 wk was significantly less statistically than the 90-min recording. From 1 day up to 8 wk only small changes were noted. Apical leakage (leakage from outside of the canal to the inside) showed a similar tendency, i.e. there was a marked decrease after the first 90min period with little change up to 8 wk. Significant difference in the apical was only seen between the 90-rain recording and the 8-wk recording. At any test period coronal leakage was greater than apical leakage. However, this difference was only significant for the 90-rain readings. Gaps were noticed between the amalgam and tooth structure throughout their interface (Fig. 2). These gaps were recorded as wide as 20 urn. There was no obvious correlation between microleakage and the width of the gap appearing on the surface.
Several studies have reported the sealing ability of retrograde amalgam fillings. Some ofthese studies were made after short time periods of 2 to 38 days (2-6). Others were made after relatively long time periods of 90 days to 22 months (13-16). Dyes and radioisotope tracers were commonly used to evaluate the leakage. But dye and radioisotope penetration studies only show the maximum distance that the tracer penetrated, i.e. linear measurement. These measurements arc not quantitative, and the results are difficult to evaluate and transfer to clinical situations. Mattison et al. (6), using an electrochemical system to measure microleakage in vitro, concluded that 3-mm deep amalgam fillings sealed better than l-ram deep fillings. They also showed that a cavity varnish improved the seal. Although their study was well-designed and gave quantitative data, it was not suitable for repeated evaluations of the same specimen over an extended time period. Kos el al. (3) used bacterial penetration to show microleakage of retrograde fillings. Their study simulated clinical situations quite well but again did
Voh 16, No. 1, January 1990
not quantitate the penetration, or allow for repeated trials over time. Other authors advocate apical fillings 2-ram deep or deeper for better seals. However, in preliminary studies using this system we found little or no leakage whcn dentin discs with standard sized amalgam plugs were tested for marginal permeability, if the discs were more than l - m m thick. It is also true that it is difficult to prepare cavities deeper than 1 mm in many clinical situations. For these reasons we used l-ram cavities in this study. Even less leakage might have resulted if we had placed thicker amalgam fillings. Both coronal leakage and apical leakage decreased markedly after 90 rain, and this change continued up to 6 h after amalgam filling. No previous leakage studies have examined this time period. This rapid decrcase in microleakage might be caused by setting expansion of amalgam. This suggests that the amalgam properties themselves as well as amalgam manipulation should be studied in greater detail for their effect in marginal sealing. Studies by Nelson and Mahler (17) using air pressure to test amalgam fillings 2 mm in diameter placed in ceramic discs 2-ram thick showed that amalgams differed in their ability to seal. They also showed that all zinccontaining amalgams contaminated with water exhibited postsetting expansion. However, only low copper alloys containing zinc exhibited delayed large expansion that could lead to root fracture. Further studies using water pressure a n d / o r air pressure to test the marginal seal of root canal fillings placed in vivo, but examined after extraction, would be most helpful. Perhaps even better seals would have resulted in our study if a cavity liner had been placed before the amalgam. It would be particularly interesting to see if these liners were dissolved over time and to note any changes in the seal due to this event. Although large changes in the rate of leakage were not shown after the first day, apical leakage tended to improve or remain unchanged in the 8 wk postfiUing. This increased sealing ability may be due to corrosion products mentioned in earlicr reports (7, 8). Kimura (14) concluded that nonzinc amalgam showed slightly less marginal penetration to dye in dog's teeth extracted 22 months after filling than did zinccontaining amalgam. The specimens in our study were stored only in water, and thus the corrosion may have been less and the seal less than if the specimens had contacted saliva or blood. Studies continue to examine this possibility and to examine even longer postfilling times. Postfilling leakage from coronal pressure was greater than apical pressure leakage, at any time period, although the differences were not statistically significant except at 90 min after filling. The apical pressure could have pushed the amalgam into the canal and reduccd the leakage, while the coronal pressure would tend to displace the amalgam out of the canal and increase the leakage, as wc recorded, After 1 day, amalgam expansion may have reduced voids at the amalgamdentin interface so that the prcssure direction would not make a difference. Chloroform and instruments were used to remove the gutta-percha and sealer coronally after the amalgams initial set. The possibility exists that some root canal filling remained and effected the results. However, the similarity between the overall external and internal test results suggests otherwise. Scanning electron microscopy showed noticeable voids between amalgam and tooth. These voids were as wide as 20
Retrograde Amalgam Seals
, m . These findings agree with former investigators who reported gaps ranging between 10 and 45/am (18, 19). We could not agree with Stabholz et al. (19) that microleakage correlated with the width of the voids observed superficially. Kaplan et al. (20) state that the clinical success of retrograde amalgam fillings may not be due to the fillings absolute seal but be duc to: (a) the fact that the apicoectomy itself removes the uncleaned unfilled lateral canals or uncleaned unfilled portions o f the main canal; (b) the fact that the apical preparation cleans the main canal, or (c) the fact that apical curettage may stimulate the healing process. It may also be true that this combined therapy (amalgam filling and surgery) reduces the inflammatory drive to a point where healing seems to have taken place even though the entire canal system is neither completely cleaned nor completely sealed and even though subclinical inflammation continues. The measurement system used in this study has shown that microleakage can be measured quantitatively and longitudinally using the same teeth and that the teeth can also be tested for leakage from a coronal direction as well as from an apical direction. The ability to test root canal seals from a coronal direction now allows us to test the clinical hypothesis that some root canal fillings, with or without apical amalgams, fail because the seal of the coronal restorations fail. This should be studied. This study presented a useful experimental system that is capable of quantitation of apical seal leakage. We would like to thank Dr. W. Fahrenbach for his help in the scanning electron microscope section and Dr. Ying F. Ma for his help in statistical analysis. Dr. Yoshimura rs affiliated with the Department of Operative Dentistry, School of Dentistry, Hokkaido University, Sapporo, Japan.Dr. Marshall is chairman and professor, Department of Endodontology, School of Dentistry, Oregon Health Sciences University, Portland, OR. Dr. Tinkle ~s assistant I~ofessor, Department of Endodontology, School of Dentistry, Oregon Health Sciences University. Address requests for reprints to Dr. F. James Marshall. Department of Endodontology, School of Dentistry, Oregon Health Sciences University, 61' 1 SW Campus Drive, Portland, OR 97201.
References 1. Farrar JN. Radical treatment of alveolar abscess. Dent Cosmos 1880:22:376-83 2. Barry GN, Selbst AG, D'Anton EW, Madden RM. Sealing quality of polycarboxylate cements when compared to amalgam as retrofilling material. Oral Surg 1976;42:109-16. 3. Kos WL, Aulozzi DP, Gerstein H A comparative bacterial microleakage study of retrofilling materials. J Endodon 1982;8:355-8. 4 Abdal AK, Retief DH, Jamlson HC The apical seal via the retrosurgical apwoach Oral Surg 1982:54:213-8. 5. Szeremeta-Browar TL, Vancura JE, Zaki AE. A comparison of the sealing properties of different retrograde techniques: an autoradiographic study. Oral Surg 1985:59:82-7. 6. Mattlson GO, yon Fraunhofer JA, Delivanis PD, Anderson AN. Microleakage of retrograde amalgams. J Endodon 1985:11:340-5. 7 Phillips RW, Gilmore HW, Swartz ML. Schenker SI Adaptation of restorations in vrvo as assessed by "~Ca J Am Dent Assoc 1961 ;62:23-34. 8. Smith GA, Wilson NHF, Combe EC. Microleakage of conventional and ternary amalgam restorations in vitro. Br Dent J 1978;144:69-73. 9. Derkson GD, Pashley DH, Derkson ME. Micro~eakage measurement of selected restorative mateials: a new in vitro method. J Prosthet Dent 1986;56:435.-40. t 0. Anderson RW, Powen BJ, Pashley DH. Microleakage of three temporary endodontic restorations. J Endoclon 1988; 14:497-501. 11. Morgan LF, Mongomery S. An evaluation of the crown-down pressuretess technique. J Endodon 1984;10:491-8. 12. Endodontic laboratory manual. The Oregon Health Sciences University, School of Dentistry, Department of Endodontology, 1987. 13. Delivanis P, Tabibi A. A comparative seafabdity study of different retrofilling materials. Oral Surg 1978;45:273-81
Yoshimura et al.
14 K~muraJT. A comparative analysis of zinc and non-zinc alloys used in retrograde endodont~c surgery. Part 1; apical seal and tissue reaction. J Endodon 1982;8:359-63 15 TrOnstad L, Trope M, Doering A, Hasselgren G Seahng ability of dental amatgams as retrograOe filltngs in endodontic therapy. J Endodon 1983;9: 551-3 16. Bramwell JD, Hicks ML. $eahng ability of four retrofilling techniques. J Endodon 1986.12:95-100. 17. Nelson LW, Mahler DB. Factors influencing the seahng behavior of
Journal of Endodontics retrograde amalgam fillings Oral Surg (in press) 18. Tanzlll~ JP Raphael D. Moodn~k RM A comparisOn of the marginal adaptation of retrograde techniques: a scanning electron microscopic s~udy Oral Surg 1980:50:74-80 19. Stabholz A. Shani J, Friedman S, Abed J. Margfnal adaptation of retrograde fillings and its correlation with sealability J Endodon 1985:11: 218-23 20. Kaplan SD, Tanzilh JP. Raphael D. Moodnik RM A companson of the marginal leakage of retrograde techniques. Oral Surg 1982;54:583-5