Release and detection of dental corrosion products in vivo: Development of an experimental model in rabbits Norbert Reuling, W. Wisser, and A. Jung Department of Prosthodontic Dentistry, Dental School, University of Marburg, Georg-Voigt-Strape 3, 0-3550 Marburg, F.R. G.
H. 0. Denschlag Department of Nicleav Chemistry, University M a i m Fritz-Strassmann-Weg 2 , D-6500 Mainz, F.R.G. An experimental animal model was developed to investigate the release of metal ions from nonprecious dental alloys. Cast specimens of five Ni-Cr-alloys and Co-Cralloys were implanted intramuscularly in rabbits for periods of 2, 4, 8, and 12 weeks. The concentrations of nickel, chromium, cobalt, and molybdenum in the implantloaded muscles were determined by electrothermal atomic absorption spectrometry (AAS) and neutron activation analysis (NAA). Reference muscle samples of each animal were analyzed to determine the individual control values. S i w c a n t increases
in the tissue concentrations of these metals occurred in the immediate vicinity of the implants. Concentration gradients of the corrosion products were found between the implant contact tissue and the implant periphery ( p < 0.001). Tissue concentrations of nickel and chromium correlated ( r < 0.7). Microprobe analysis before and after implantation of the alloy specimens indicated an even corrosive loss of the alloy surfaces and changes in the surface element distributions. Advantages and limitations of this animal model are discussed, as well as its application in future studies.
INTRODUCTION
The biocompatibility and the corrosion behavior of dental alloys are closely related.'-3 Corrosion of alloys is a precondition for nearly all their adverse biological effects in the oral ~ a v i t y .Therefore, ~-~ investigation of corrosion processes and detection of dental corrosion products in vivo have special significance in the context of biocompatibility studies of dental all o y ~ . ~Recently ,' many corrosion studies have been performed in vitru using anodic polarization techniques, as well as other electrochemical The results of those investigations performed in vitro may differ considerably from the corrosion behavior of dental alloys in the oral cavity.'f3The situation with respect to corrosion in vivo, especially as it pertains to and is influenced by the physiological environment of the oral cavity, is complex. It is characterized by multiple interactions among many contributing variables. These variables include pH, temperature, physical stress, oxygen concentration, presence of proteins and microbiologic a~tivities.'-~,~ Journal of Biomedical Materials Research, Vol. 24, 979-991 (1990) 0 1990 John Wiley & Sons, Inc. CCC 0021-9304/90/080979-13$04.00
REULING ET AL.
980
Experimental ~ f u d i e s ~ revealed , ’ ~ ’ ~ a mutual dependence of proteins and oral microorganisms (bacteria, fungi) in the development of biological corrosion processes. Thus, in humans biodegradation of metallic biomaterials, especially dental alloys, caused by highly corrosive body fluids in physiological environments like the oral cavity, may disturb the local and systemic balance among trace elements.’ Moreover, dental corrosion products may produce local or systemic toxicological effects.20-22 The induction or provocation of immunologic responses, like hypersensitivity due to alloy constituents, for example nickel, chromium, cobalt, or gold, is well recogni~ed.’~-= Finally, very few studies have been performed in man concerning carcinogenic risks and mutagenic effects of dental corrosion p r o d u ~ t s . ’ ~ , ~ ~ To elucidate adverse biological effects of dental alloys and their corrosion products, trace element analysis of tissues and body fluids of humans and experimental animals is critical because such analysis provides more infor,mation about the local and systemic influences and interactions of corrosion products and their biological pathways in the ~rganism.~,’ Several experimental studies have shown that microtrace element analysis techniques allow quantitative detection of metallic elements in various biological matrices, e.g., blood, urine, saliva and t i s ~ u e s . ~ ,For ” ~ all ~ - these ~ ~ techniques it is particularly important to avoid metallic contamination of the biological specimens during experimental procedures.8 The aim of the present study was to develop an animal model that would permit establishment of methodical and experimental principles for the detection of corrosion products released from nonprecious dental alloys. EXPERIMENTAL
Materials Five nonprecious dental casting alloys were used. Four of these alloys contain mainly nickel and chromium, the fifth principally cobalt and chromium. The compositions of the alloys are summarized in Table I. All alloys were cast in accordance with directions provided by the manufacturers. The implant design and dimensions were in accordance with a recently described implant model3’ (Fig. 1). The surfaces of the implants were treated TABLE I Chemical Composition of the Nonprecious Dental Alloys Investigated Composition (wt %) ~
Material
Ni
Ni-Cr alloy I Ni-Cr alloy I1 Ni-Cr alloy 111 Ni-Cralloy IV Co-Cr alloy I
64.0 64.1
64.0 76.0
Co
Cr
Mo 10.0
0.5 69.5
24.0 22.0 22.0 13.0 24.0
9.0 9.0 3.0 4.5
Ti
~
Be
Ga
~~
Fe
1.0
2.0
2.0
1.5
1.0
Nb
Si
3.7
0.5
4.0 1.0
IN VIVO DETECTION OF DENTAL CORROSION PRODUCTS
981
Imml
Figure 1. Design and dimensions of the used implant type (see ref. 32).
after casting with abrasives like carborundum stones and afterward polished to a 6-pm finish by use of dental polishing paste which contains ferric oxide, respectively chromium oxide. Finally the implants were ultrasonically cleaned in 98% ethanol and afterward steam sterilized.
Experimental procedure In 20 male albino rabbits (New Zealand strain, body weight: 2.2-2.4 kg) 120 implants of five nonprecious dental alloys were implanted into the left paravertebral muscles for periods of 2, 4, 8, and 12 weeks.* A total of 960 muscle samples were removed from animals sacrificed after the specified implant periods. Tissue concentrations of the alloy elements nickel, chromium, cobalt, and molybdenum were determined quantitatively by means of electrothermal atomic absorption spectrometry (moist chemical and solid sample analysis) and by means of neutron activation analysis. Control muscle samples taken from the right unimplanted paravertebral muscle of each rabbit were analyzed to determine individual control values of the metal elements. Because it was particularly important to avoid metal *The national regulations of the Federal Republic of Germany for the care and use of laboratory animals have been observed.
982
REULING ET AL.
contamination of the muscles and the removed tissue samples during all experimental and analytical parts of our study, only surgical and analytical instruments and materials made of nonmetallic constituents were used, e.g., ceramics, plastics, Teflon, as well as metallic scissors and scalpels coated with titanium-nitride. Furthermore, only highly pure chemical agents (double distilled acids and Aqua bidest.) and trace element-free laboratory vessels were used.
Implantation The implantation of the alloy specimens was performed under general anesthesia. The back of each rabbit was shaved and the skin disinfected. All surgical procedures were carried out under aseptic conditions. A skin incision on the left paravertebral side was made with a titanium-nitride-coated scissor. The subcutaneous tissue layer was bluntly dissected and six incisions into the left paravertebral muscle were made with a ceramic scalpel. In each rabbit six specimens of the same alloy were carefully implanted intramuscularly after blunt preparation of six tissue pockets. Finally the skin was closed with sutures. The whole surgical procedure was done exclusively with ceramic instruments and with titanium-nitride-coated scissors.
Dissection of tissue samples After each implantation period both paravertebral muscles of each animal were carefully dissected. Again only titanium-nitride-coated scissors and scalpels were used to avoid metal contamination of the tissue. The right paravertebral muscle of each rabbit not containing alloy implants was used to determine individual control values of the metal elements nickel, chromium, cobalt and molybdenum. The implant-loaded muscles were cut into smaller pieces of equal size. Then each implant was carefully removed from the muscle sample without damaging or removing the fibrous capsule surrounding the implants.
Tissue sample storage The muscle samples were put in polyethylene laboratory vessels which had been cleaned by leaching with double-distilled, highly pure nitric acid and Aqua bidest. Afterward the samples were immediately shock frozen in liquid nitrogen. The frozen biological material was stored at -20°C for less than 1 week.
Trace element analytical aspects The tissue concentrations of the metal elements nickel, chromium, cobalt, and molybdenum in implant-loaded muscle samples and in reference
IN VlVO DETECTION OF DENTAL CORROSION PRODUCTS
983
muscle samples were determined quantitatively by three analytical techniques: moist chemical atomic absorption spectrometry (electrothermal), solid matter atomic absorption spectrometry (electrothermal), and neutron activation analysis. All analytical data refer to values in ppb wet weight (ng/g). The moist chemical AAS-analysis was performed (Perkin-Elmer model 4000) using pyrolized graphite furnace tubes with an electrothermal source (Perkin-Elmer HGA 400) and the Zeemann-compensation technique. The solid matter AAS-analysis was carried out with the Zeemann-AASequipment SM 20 (Grun Optik Company, Wetzlar, FRG). To perform direct analysis of solid tissue samples with atomic absorption spectrometry, very small muscle pieces were removed from freeze-dried muscle samples with ceramic forceps. Individual tissue samples were weighed electronically and a so-called platform-boat was used as a device to insert the solid sample into the electrothermal atomizer. From each sample microgram specimens were removed from the immediate vicinity of the implant, as well as from tissue areas at a distance of 1 cm and of 2 cm away from the implant. From each animal we removed several muscle samples of the right paravertebral muscle unloaded with implants and we determined the individual control values of the tissue concentration of the metal elements by use of the above-named analytical methods. To perform neutron activation analysis of the biologic specimens the irradiation facilities of the research reactor of the University Mainz were used. The irradiation of the muscle samples was done with thermal neutron fluxes of 7 X 10'' n X cm-* x s-' for 2 h. After irradiation, radiochemical separation of the samples was necessary because an exact gamma-spectrometric detection and quantitative analysis of the selected elements, especially nickel, could not be attained otherwise. This is due to superimposition of highenergy beta- and gamma-radiation of essential elements, e. g., sodium, potassium, and chloride. For this reason several combined precipitations with iron chloride and ammonium sulfide were carried out after a nitric acid/perchloric acid separation. In this way separation of the interfering anorganic elements, as well as of the entire organic matrix, was achieved. Afterward, gamma-spectrometricanalysis of the alloy elements could be done.
Surface analytical aspects Before and after each implantation period, precisely defined surface areas of the implants were examined by means of energy dispersive microprobe analysis. By this means the distribution of the elements nickel, chromium, cobalt, and molybdenum on the implant surface was determined. RESULTS
Significant increases above control values occurred in the tissue concentrations of nickel, chromium, cobalt, and molybdenum for all alloys exam-
REULING ET AL.
984
Contact tissue
1L
30
60
90
Implantation period [days]
Figure 2. Nickel tissue concentrations (mean values) in the vicinity of alloy implants (Ni-Cr alloy I) determined by means of solid matter atomic absorption spectrometry.
ined. Figures 2 and 3 graphically show the mean nickel and chromium concentrations of the Ni-Cr-alloy I measured by means of solid matter-AAS. Table I1 and 111 present the mean values (x) and the standard deviations (s) of the nickel concentrations respectively the chromium concentrations of the Ni-Cr-alloy I. In the contact tissue nickel concentrations reached 715 ng/g after 4 weeks. Tissue concentrations at a distance of 1 cm away from the implant also showed markedly elevated nickel concentrations
I
-P 700 600
I
s 500
c
g Loo 01
g
3M3
V
01
l= nI
200
Contact tissue
,% 100
+
0 14
30
60
90
Implantation period [days]
Figure 3. Chromium tissue concentrations (mean values) in the vicinity of alloy implants (Ni-Cr alloy I) determined by means of solid matter atomic absorption spectrometry.
iN
vivo DETECTION OF DENTAL CORROSION PRODUCTS
985
compared to the control values. The tissue concentration at a distance of 2 cm from the implant exhibited only slight accumulation of the element nickel. Chromium tissue concentrations (Fig. 3 and Table 111) are significantly increased in the implant contact tissue. Similar to the results for nickel a concentration gradient occurred for chromium from tissue areas nearest the implant to the implant periphery. Also at a distance of 2 cm from the implant chromium levels were distinctly raised above control values. The highest concentrations of nickel and chromium were found for the Ni-Cr-alloy 111. Figure 4 shows that the mean nickel concentrations in the
TABLE I1 Nickel Tissue Concentrations Determined by Means of Solid Matter ASS (Ni-Cr Alloy I) Nickel (ng/g)
x
Blank values
S
-
Concentration in 2-cm distance
X S
x
Concentration in l-cm distance
S
-
Concentration contact tissue
X S
14 Days
30 Days
60 Days
90 Days
47.00 1.43
18.20 1.86
19.60 4.77
39.70 24.10
82.10 70 20 I
11.40 1.32
16.30 4.02
28.30 20.10
45.50 13.90
36.50 17.00
42.80 12.00
99.70 24.40
137.00 35.10
715.00 423.00
192.00 75.50
448.00 81.50
TABLE 111 Chromium Tissue Concentrations Determined by Means of Solid Matter ASS (Ni-Cr Alloy I) Chromium (ng/g) Blank values
-
X
S
Concentration in 2-cm distance Concentration in 1-cm distance Concentration contact tissue
X S
Tf S
-
X S
60 Days
90 Days
14 Days
30 Days
25.90 17.60
9.68 7.21
5.45 1.05
6.42 3.67
19.70 1.56
37.20 18.60
27.90 6.49
22.20 6.65
83.00 27.60
30.90 4.33
27.90 1.77
28.70 5.47
133.00 57.50
71.40 26.30
408.00 320.00
713.00 560.00
986
REULING ET AL. I
Y
14
Contact tisue
30 M) Implantation period Idays]
90
Figure 4. Nickel tissue concentrations (mean values) in the vicinity of alloy implants (Ni-Cr alloy 111) determined by means of solid matter atomic absorption spectrometry.
implant contact tissue reached more than 3000 ng/g after an implantation period of 2 weeks. Table IV presents all mean values ( x ) and standard deviations (s) of the nickel concentrations of the Ni-Cr-alloy 111. The Ni-Cr-alloy I11 showed the highest nickel concentrations in the contact tissue of all NiCr-alloys examined. These results were confirmed by those of the moist chemical AAS-analysis and the neutron activation analysis. The trace element analysis of tissues which had been in contact with the Co-Cr-alloys I showed very high cobalt concentrations in the near implant vicinity, reaching 726 ng/g. A distinct gradient of the cobalt concentration from the implant contact tissue to the tissue at distances of 1and 2 cm from the implant appeared in all specimens examined. The moist chemical AASresults and the results of the neutron activation analysis showed distinctly TABLE IV Nickel Tissue Concentrations Determined by Means of Solid Matter ASS (Ni-Cr Alloy 111) Nickel (ng/g) Blank values
-
X
S
Concentration in 2-cm distance Concentration in 1-cm distance Concentration contact tissue
X S
fl S
-
X s
14 Days
30 Days
60 Days
YO Days
54.80 32.70
38.70 13.40
12.80 3.89
33.80 8.82
18.20 9.98
13.90 5.50
16.90 8.77
35.50 13.40
25.10 1.86
9.07 3.52
40.20 6.75
18.90 6.23
3115.00 896.00
1842.00 206.00
662.00 242.00
765.00 409.00
IN vrvo DETECTION OF DENTAL CORROSION PRODUCTS
987
raised cobalt and chromium concentrations in the implant-loaded muscle tissue compared to the individual control values of these elements. The energy dispersive microprobe analysis of the alloys showed changes in the element distribution on the alloy surfaces. Those changes indicate that a general corrosion process has taken place. No signs of pitting corrosion were found. Statistical analysis of the results revealed the following: 1)Significantly increased concentrations of nickel, chromium, cobalt, and molybdenum were found in implant-loaded tissues. Mean nickel concentrations of 653 ng/g (SD: -+ 89 ng/g) and mean chromium concentrations of 493 ng/g (SD: k 71 ng/g) were present in the implant contact tissue as compared to mean control values of 32.4 ng/g (SD: -+ 4.7 ng/g) for nickel and 27.1 ng/g (SD: 2 3.4 ng/g) for chromium. 2) By means of analysis of variance, statistically significant differences were found between the nickel and chromium concentrations in the contact tissue and the control values, respectively, at distances of 1 cm and 2 cm away from the implanted alloys ( p < 0.001). 3) Statistical rank correlation calculations (Spearman-test) revealed a mean correlation ( Y < 0.7) of the tissue concentrations of the elements nickel and chromium. No significant correlation between the tissue concentrations of cobalt and chromium was observed. 4) The corrosion behavior of the Ni-Cr-alloys I, 11, and IV differed significantly from the corrosion behavior of the Ni-Cr alloy 111.
DISCUSSION
Because of methodical and ethical limitations of human corrosion studies, an experimental rabbit model was developed to study the corrosion behavior of dental alloys. To examine the corrosion behavior of these alloys, microtrace element analytic techniques were utilized. Similar recent studies mainly concerned orthopedic and surgical alloy^.^,^^,^^ After intramuscular implantation of four Ni-Cr alloys and one Co-Cr alloy, very high enrichments occurred of the alloy elements nickel, chromium, and cobalt in the peri-implant tissues. In all alloys there was a gradient of metal elements released, declining progressively from the nearest implant tissue areas to the implant periphery. These results are in agreement with those of other investigators who observed similar concentration gradients of metal elements in human tissues after implantation of orthopedic alloys.34Also in agreement with other authors,35we observed no dependence of the length of the implantation period and the corrosion behavior of the implanted alloys. We observed significant differences between the Ni-Cr alloys I, 11, IV, on one hand, and the Ni-Cr alloy 111, on the other hand. These differences cannot be explained totally, because the elemental composition of the Ni-Cr alloy I11 differs negligibly from that of the Ni-Cr alloys I and 11. Also the ele-
988
REULING ET AL.
ment distribution maps recorded by help of energy dispersive microprqbe analysis, showed no differences concerning the microstructure or the surface element distribution between the four Ni-Cr alloys examined. Although the Ni-Cr alloy IV contains the elements beryllium and gallium and furthermore has a very low chromium content, we could not prove increased corrosion rates of this alloy. This contrasts to previous results obtained in vitro.36 A statistical rank correlation analysis showed a mean correlation of the tissue concentrations of nickel and chromium in all Ni-Cr alloys examined in our study. This relationship suggests that an increase in corrosive release of nickel always is accompanied by a release of chromium and vice versa. The corresponding ratio of the tissue concentrations of nickel and chromium did not agree with the weight percent relation of these elements in the alloys. These results confirm other corrosion studies concerning surgical Co-Cr alloys and orthopedic Ni-Cr ~ t e e l s . Until ~~-~ now it is unknown if the constant tissue ratios of released metals are caused by biological carrier systems of the specific metal elements, or metabolic processes, or both.41 These results concerning the Co-Cr alloy bear a strong resemblance to those of other authors who found high cobalt concentrations in the connective tissue around implanted dental Co-Cr alloy specimens.3o The results of the energy dispersive microprobe analysis confirmed distinct changes of the element distribution on the alloy surface which indicate an even corrosive wear process. Other experimental studies proved similar results by use of comparative trace element analysis and electron spectroscopy of high-grade steel implantsm The comparison of the advantages and disadvantages of these analytical methods shows that solid matter atomic absorption spectrometry allows locally defined removal and analysis of very small tissue samples. This made it possible to evaluate tissue concentration gradients of corrosive released metal elements. Neutron activations analysis is a multielement technique which makes it possible to determine multiple elements simultaneously. Further advantages of this radiochemical method are its inherent low blank values, its excellent accuracy, and the extremely low detection limits. Disadvantages of neutron activation analysis are the high methodical efforts, the radiation hazards, and the removal of radioactively contaminated laboratory equipment and biological samples. Some aspects of this study deserve special critical comments. The corrosion model used alloy implants lying passively in the muscle tissue of rabbits while in clinical use prosthodontic appliances made of those dental alloys are subjected to the intraoral environment with various physical stresses, e.g., occlusal wear during mastication. Our alloy implants were exposed to the internal environment of an animal, namely the extracellular fluid of rabbit muscle tissue. In contrast to the oral environment, the extracellular fluid shows biochemical stability due to efficient regulatory mechanisms. Therefore, the main corrosion parameters, e.g., pH, oxygen concentration, temperature, and concentration of inorganic cations like chlorides and phosphates, remain in narrow ranges,
IN VIVO DETECTION OF DENTAL CORROSION PRODUCTS
989
The oral environment has much more variability. The corrosive medium of the oral cavity is saliva. The physiological saliva composition differs between individuals as a function of general diseases, drug administration, and nutritional habits of humans. Substantial changes in the oral environment appear because of oral ingestion of foods and drinks which may contain numerous corrosive components and often cause extreme fluctuations of the temperature in the oral cavity. Therefore extrapolations of these experimental corrosion results to the intraoral corrosion behavior of these alloys in humans should be limited. This study was funded in part by the Deutsche Forschungsgemeinschaft (DFG) and the Deutsche Gesellschaft f i r Zahn-, Mund- und Kieferheilkunde (DGZMK). We gratefully acknowledge the support of the Griin Optik Company, Wetzlar, FRG, especially for the use of their solid matter atomic absorption spectrometer SM 20.
References 1. M. Marek, “Corrosion in a biological environment,” in International
2.
3.
4.
5.
6. 7. 8. 9.
10. 11. 12. 13.
14.
Workshop on Biocompatibility, Toxicity and Hypersensitivity to Alloy Systems Used in Dentistry, B. R. Lang, H. F. Morris, and M. E. Razzoog (eds.), University of Michigan, Ann Arbor, 1985, pp. 103-122. D. F. Williams, “Physiological and microbiological corrosion,” CRC Crit. Rev. Biocompat., 1, 1-24 (1985). W. T. Klotzer, “Biologische Aspektc der Korrosion,” Dtsck. Zahniirztl. Z., 40, 1141-1145 (1985). K. Merritt, “Biochemistry, hypersensitivity, clinical reaction,” in International Workshop on Biocompatibility, Toxicity and Hypersensitivity to Alloy Systems Used in Dentistry, B. R. Lang, H. F. Morris, and M. E. Razzoog (eds.), University of Michigan, Ann Arbor, 1985, pp. 195-223. D. Brune, ”Metal release from dental biomaterials,” Biornaterials, 7, 163175 (1986). D. Brune, A. Kjaerheim, A. Hensten-Pettersen, and L. Marion, “Corrosion of dental alloys studied by nuclear tracer technique,” Acta Odontol. Scand., 41, 129-134 (1983). R. Soremark, “Some biological effects caused by prosthetic materials,” Swed. Dent. I,, 3, 1-7 (1979). R. Michel, ”Trace metal analysis in biocompatibility testing,” CRC Crit. Rev. Biocornpat., 3, 235-317 (1987). M. Pourbaix, “Electrochemical corrosion of metallic biomaterials,” Biomaterials, 5, 122-135 (1984). S. M. deMicheli and 0. Riesgo, ”Electrochemical study of corrosion in Ni-Cr-dental alloys,” Biomaterials, 3, 209-212 (1982). M. W. Espevik, “Corrosion of base metal alloys,” Acta Odontol. Scand., 36, 113-117 (1978). L. C. Lucas, R. A. Buchanan, J. E. Lemons, and C. D. Griffin, “Susceptibility of surgical cobalt-base alloy to pitting corrosion,” 7. Biomed. Mater. Res., 16, 799-810 (1982). J. M. Meyer, J. M. Wirthner, R. Barrand, C. P. Susz, and J. N. Nally, ”Corrosion studies on nickel-based casting alloys,” in Corrosion and Degradation of lmplant Materials, B. C. Syrett and A. Achrya (eds.), American Society for Testing and Materials Special publication 684, ASTM, Philadelphia, 1979, pp. 295-315. S. A. Brown, L. J. Farnsworth, K. Merritt, and T.D. Crowe, “In vitro and in vivo metal ion release,” J. Biomed. Mater. Kes., 22,321-338 (1988).
990
REULING ET AL. 15. G. C. F. Clark and D. F. Williams, “The effects of proteins on metallic corrosion,” J. Biomed. Mater. Res., 16, 125-134 (1982). 16. K. Merritt, S. A. Brown, and N. A. Sharkey, “The binding of metal salts 17. 18. 19. 20. 21.
22.
23. 24. 25. 26. 27. 28. 29. 30. 31. 32.
33.
and corrosion products to cells and proteins in vitro,” J. Biomed. Mater. Res., 18, 1005-1015 (1984). H. J. Mueller, “The binding of corroded metallic ions to salivary-type proteins,“ Biomaterials, 4, 66-72 (1983). H. J. Mueller, ”Binding of corroded ions to human saliva,” Biomaterials, 6, 146-149 (1985). G. Palaghias and R. Soremark, “The electrochemical properties of three amalgam alloys in cultures of Streptococcus mutans, ” J. Dent. Res., 63, 583, Abstr. 104 (1984). D. S. Smith, “Tissue reaction of noble and base metal alloys,” in Biocompatibility of Dental Materials, Val. IV, D. 5. Smith and D. F. Williams (eds.), CRC Press, Boca Raton, 1982, pp. 51-77. C.R. Culliton, M.A. Meenagham, S.E. Sorensen, G. W. Greene, and J. D. Eick, “A critical evaluation of the acute systemic toxicity test for dental alloys using histopathologic criteria,” J. Biomed. Mater. Res., 15, 565-575 (1981). A. Hensten-Pettersen and N. Jacobsen, “Biocompatibilityof dental base metal alloys as evaluated by subcutaneous implants in rats and by cell culture technique,” in Evaluation of Biomaterials, G. D. Winter, J. L. Leray, and K. deGroot (eds.), Wiley, London, 1980, pp. 44-447. J. Black, ”Systemic effects of biomaterials,” Biomaterials, 5, 11-18 (1984). B. Magnusson, M. Bergman, B. Bergman, and R. Soremark, ”Nickel allergy and nickel containing alloys,” Scand. J. Dent. Res., 90, 163-167 (1982). K. Merritt and S. A. Brown, ”Hypersensitivity to metallic biomaterials,” in Systemic Aspects of Biocompatibility, Vol. 11, D. F. Williams (ed.), CRC Press, Boca Raton, 1981. J. Autian, “Carcinogenic potential of metals,” in Workshop on Biocornpatibility of Metals in Dentistry, American Dental Association, Chicago, 1984, pp. 107-113. C. P. Flessel, A. Furst, and 5. B. Radding, “A comparison of carcinogenic metals,” in Metal Ions in Biological Systems, Vol. X , H. Sigel (ed.), Marcel Dekker, New York, 1979. F. Fontes de Melo, N. R. Gjerdet, and E. S. Erichsen, “Metal release from cobalt-chromium partial dentures in the mouth,” Acta Odontol. Scund., 41, 71-74 (1983). R. Michel, J. Hofmann, F. Loer, and J. Zilkens, “Trace element burdening of human tissues due to corrosion of hip-joint prostheses made of cobalt-chromium alloys,” Arch Orthop. Trauma Surg., 103, 85-90 (1984). T. Stenberg and B. Bergman, “Release and uptake of cobalt from cobaltchromium alloy implants,” Actu Odontol. Scund., 41, 149-154 (1983). M. Bergman, B. Bergman, and R. Soremark, ”Tissue accumulation of nickel released due to electrochemical corrosion of non-precious dental casting alloys,” J. Oral Rehabif., 7, 325-330 (1980). V. Geret, B.A. Rahn, R. Mathys, F. Straumann, and S.M. Perren, ”A method for testing tissue tolerance for improved quantitative evaluation through reduction of relative motion at the implant-tissue interface,” in Evaluation of Biomaterials, G . D. Winter, J. L. Leray, and K. deGroot (eds.), Wiley, London, 1980, pp. 351-359. R. Michel, F. Loer, M. Nolte, M. Reich, and J. Zilkens, ”Phenomenology of the trace element burdening of the human organism by the in body corrosion of Co-Cr-Ni-alloys as revealed by neutron activation analysis,” in Biocompatibility of Co-Cr-Ni-Alloys, H. F. Hildebrand and M. Champy (eds.), NATO-AS1 Series, Plenum Press, New York, 1987.
IN VIVO DETECTION OF DENTAL CORROSION PRODUCTS
991
34. J. Schuster, F. Lux, and R. Zeisler, “Untersuchungen der Metallose durch Neutronenaktivierungsanalyse,” Msckr. Llnfallheilk, 76, 537-548 (1973). 35. R. Michel and J. Zilkens, “Untersuchungen zum Verhalten von Metall36. 37.
38. 39. 40.
41.
42.
43. 44. 45.
46.
spuren im umgebenden Gewebe von AO-Winkelplatten mit Hilfe der Neutronenaktivierungsanalyse,” Z . Ortkop., 116, 666-674 (1978). J. Covington, “Preliminary quantization of salivary nickel leakage from non-precious alloys,” J. Tenn. Dent. Ass., 62, 16-19 (1982). M. Abeln, J. Ohnsorge, and K. Kasperek, “Biochemische Untersuchungen der legierungsspezifischen Spurenelemente von KobaltChrom-Huftgelenktotalendo-prothesen und AO-Winkelplatten im implantatnahen und implantatfernen Gewebe sowie im Serum mit Hilfe der Neutronenaktivierungsanalyse,” in Grenzsckichtprobleme der Verankerung von Implantaten unter besonderer Berucksichtigung von Endoprothesen, M. Jager, M. H. Hackenbroch and H. J. Refior (eds.), Thieme, Stuttgart, 1980. F. Loer, J. Zilkens, Hofmann, and R. Michel, ”Zum Nachweis korperfremder Spurenelemente nach Langzeitimplantation von Totalendoprothesen aus Kobaltbasislegierungen,” Z . Orthop., 119, 763-766 (1981). H. S. Dobbs and M. H. Minski, ”Metal ion release after total hip replacement,” Biomaterials, 1, 193-198 (1980). R. Michel, J. Hofmann, R. Holm, and J. Zilkens, “Zum ijbertritt von Korrosionsprodukten aus Stahlimplantaten in das Kontaktgewebe. Untersuchungen der Implantatoberflache mit ESCA und instrumenteller Neutronenaktivierungsanalyse des Kontaktgewebes,” Z . Ortkop., 118, 793-803 (1980). F. Lux, J. Schuster, and R. Zeisler, “A mechanistic model for the metabolism of corrosion products and of biological trace elements in metallosis tissue based on results obtained by activation analysis,” 1. Radioanal. Chem., 69, 381-400 (1982). E. Sabbioni, R. Pietra, and E. Marafante, ”Metal metabolism in laboratow animals and human tissues as investigated bv neutron activation ” analysis: current status and perspectives,” J. Radioanal. Ckem., 69, 381400 (19821. F. Lux, S.’ Trebert-Haeberlin, and W. Erhardt, ”Neutronenaktivierungsanalytische Bestimmung des Chromgehaltes von Laboratoriumskaninchen-Skelettmuskel,” Fresenius Z . Anal. Ckem., 323, 833-838 (1986). A. Rosopulo, ”Schwermetallbestimmung direkt aus dem Feststoff und nach chemischem Aufschlup -ein Methodenvergleich,” Fresenius Z. Anal. Chem., 322, 669-672 (1985). B. Klupendorf, A. Rosopulo, and W. Kreutzer, ”Untersuchungen zur Verteilung und Schnellbestimmung von Blei, Cadmium und Zink in Lebern von Schlachtschweinen mittels Feststoff-ZeemannAtomabsorptionsspektrometrie,” Fresenius Z . Anal. Chem., 322, 721-727 (1985). T. Hadeishi and R. McLaughlin, ”Direct ZAAS-analysis of solid samples: early development,” Fresenius Z . Anal. Ckem., 322, 657-659 (1985).
Received January 1989 Accepted October 31, 1989
k.