T h e e f f e c t s o f s o d i u m f l u o r i d e a n d s t a n n o u s f l u o r i d e on t h e surface roughness of intraoral magnet systems R o b e r t M. O b a t a k e , D D S , a S t e p h e n M. C o l l a r d , D D S , b J a c k M a r t i n , D D S , M S , c a n d G. D a v i d L a d d , B S d

The University of Texas M.D. Anderson Cancer Center, and the University of Texas Health Science Center, Dental Branch, Houston, Texas Four t y p e s of intraoral m a g n e t s u s e d for retention of o v e r d e n t u r e s and m a x i l l o f a cial p r o s t h e s e s w e r e e x p o s e d in vitro to SnF2 and N a F to d e t e r m i n e the effects o f fluoride r i n s e s on surface roughness. The surface r o u g h n e s s (Ra) w a s m e a s u r e d , after s i m u l a t e d 1, 2, and 5 years' clinical e x p o s u r e to fluoride (31, 62, and 155 hours). The m e a n c h a n g e in Rawas calculated for each period of s i m u l a t e d exposure to fluoride for each m a g n e t type. T w o - w a y ANOVA w a s u s e d to c o m p a r e m e a n c h a n g e in Ra b e t w e e n m a g n e t s w i t h i n fluorides, and b e t w e e n fluorides w i t h i n m a g n e t s . Paired t tests w e r e u s e d to c o m p a r e m e a n c h a n g e in Ra w i t h i n fluorides w i t h i n m a g n e t s . The m e a n c h a n g e in l{ a i n c r e a s e d for all m a g n e t s after s i m u l a t e d 1, 2, and 5 y e a r s o f e x p o s u r e to SnF2 and N a F (iv < 0.03). U s i n g the c h a n g e in Ra as an indicator for corrosion, PdCo e n c a p s u l a t e d SmCo5 m a g n e t s and their k e e p e r s d e m o n s t r a t e d the l e a s t corrosion w i t h either fluoride. (J PROSTHET DENT 1991;66:553-8.)

M a g n e t i c attachment systems for overdentures offer significant advantages over mechanical attachment systems. They are generally easier to rebase, reline, and repair, and wear may be minimized if the magnets are plated or encapsulatedJ Magnetic attachment systems also permit the denture base to rock, slide, or even dislodge before automatically reseating. This movement spares the abutment roots from excessive lateral or rotational forces under occlusal loading. The retentive force and compactness of two particular magnet types, samarium cobalt (SmCos), and neodymium iron boron (Nd-Fe-B), have resulted in widespread use of these intraoral magnets in overdentures and maxillofacial prosthetics. During use, intraoral magnets and their opposing keepers may be subjected to 1% neutral pH sodium fluoride (NaF) or 0.4% stannous fluoride (SnF2), which aid in the maintenance of the supporting overdenture abutments. Derkson and MacEntee 2 found 0.4% SnF2 to be effective in reducing the progress of gingivitis around overdenture abutments if used daily for 1 year. Toolson and Smith 3 and Derkson and MacEntee 2 recommended placement of fluoride gels directly into the indentations opposing the abutments at least once a day. Shannon and Cronin 4 recommended that the patient wear the fluoride-carrying overdenture in place for 5 minutes before removing it and rinsing the overdenture.

aFellow, Maxillofacial Prosthodontics, Department of Dental Oncology. bAssistant Professor, Department of Oral Biomaterials. CAssociate Oncologist (Maxillofacial), Department of Dental Oncotogy. dResearch Assistant, Department of Oral Biomaterials. 10/1/25048 THE J O U R N A L OF P R O S T H E T I C D E N T I S T R Y

Intraoral magnets exposed to fluoride over an extended period of time may corrode and release corrosion by-products. The biologic effect of long-term ingestion of these corrosion by-products is not well documented, but release of metal ions through oral corrosion is known to have biologic and clinical effects including: (1) cytotoxicity, (2) tissue lesions, (3) metallic taste, (4) sensitizations, or (5) carcinogenesis. 5 Corrosion of the magnet may also result in diminished retention. This study evaluated the change in surface roughness of polished raw and encapsulated magnets after simulated 1, 2, and 5 years of exposure to fluoride solutions in vitro. The change in surface roughness is an indication of the amount of corrosion. MATERIAL

AND

METHODS

Fourteen of each of four magnet types were obtained from the respective manufacturers (Table I). The term "magnets" includes three types of intraoral magnets and one intraoral keeper, although the keeper is nonmagnetic. The 14 specimens of each magnet type were randomly divided into two groups of seven. Each group of seven magnets was embedded in cold-cured acrylic resin cylindrical blocks (Sampl-Kwik, Buehler, Ltd., Lake Bluff, Ill.) so that the functional surface of each magnet was exposed. The seven magnets were evenly spaced 3 to 4 mm apart. Initial measurements of the surface roughness of the magnets as supplied by the manufacturer revealed significant variation between the four magnet types. To provide for a similar initial roughness of each magnet type after being embedded in acrylic resin, the roughest magnets (SC and NI) were polished in a metallographic polisher (Automet, Buehler, Ltd.). Four hundred and 600 grit silicon carbide abrasive paper disks (Carbimet, Buehler, Ltd.) 553

OBATAKEET AL

Table

L Types of magnets evaluated Magnet

Product name

Composition

Field

Encapsulation

SC

Jobmaster No. 18 Jobmaster, Randallstown, Md. Jobmaster No. 24 Parkell Dyna-Magnet Parkell, Farmingdale, N.Y. Parkell Dyna-Keeper

SmCo5

Open

--

Nd-Fe-B SmCo5

Open Closed

-PdCo

--

--

NI DM DK

Table

PdCo

II. Mean change in magnet roughness (Ra)*: Two-way A N O V A - - f l u o r i d e versus magnet type Year(s)

Source

DF

SS

MS

0-1

Fluoride Magnet Interaction

0.00002 0.00767 0.00026 0.00472 0.01267

0.00002 0.00256 0.00009 0.00010

0.1972 25.9705 0.8902

Total

1 3 3 48 55

0-2

Fluoride Magnet Interaction Error Total

1 3 3 48 55

0.00065 0.02980 0.00061 0.00515 0.03620

0.00065 0.00993 0.00020 0.00011

6.0280 92.8318 1.8972

0-5

Fluoride Magnet Interaction Error Total

1 3 3 48 55

0.00096 0.08045 0.00050 0.00633 0.08824

0.00096 0.02682 0.00017 0.00013

7.2803 203.1515 1.2576

Error

F ratio

*Ra values in microns; n 7. SS, Sum of squares; MS, mean squares. =

were used in sequence for initial smoothing, followed by 6 ttm diamond polishing paste (Metadi, Buehler, Ltd.) on a cloth disk (Texmet, Buehler, Ltd.) for polishing. The other magnets (DM and DK), which were encapsulated, were smooth as supplied and were not polished. After polishing the SC and NI magnets, initial average surface roughness (Ra) measurements were made (in microns) on the functional surface of each magnet using a profilometer (Talysurf 10, Rank Taylor Hobson, Leicester, England). Three Ra measurements were made of the exposed surface of each magnet. After each measurement, the magnet was rotated 60 degrees about its center before the next measurement. The recorded Ra was the average of the three independent Ra measurements made at traverse directions 60 degrees apart. The profilometer was set at a traverse length of 0.25 m m across the center of the magnet surface with an amplification of 20,000. The meter cutoff and sampling length was set to 0.8 m m for each measurement. The stylus tip width was 0.0025 m m and the stylus tip force was 1 mN. The initial Ra values provided

554

measurements to which subsequent Ra measurements on the same magnet were compared following exposure to fluoride. Following the initial Ra measurements, the acrylic resin blocks containing the magnets were separated into two groups (each containing the four types of magnets to be tested). One of the groups was immersed in a beaker containing 0.4 % SnF2 (pH 5.5, Preventive Dentistry S u p p o r t Center, VA Medical Center, Houston, Texas), which was prepared (activated) by mixing distilled water with the water-free gel in a 1:1 ratio.* The other group was immersed in a beaker containing 1% N a F (pH 6.3, Emerson Laboratories, Texarkana, Texas). The t e m p e r a t u r e of the fluoride solutions was maintained at 37 + 1 ° C. The fluoride solutions were replaced with freshly prepared solution every hour for the duration of the experiment. Each time

*Wybourny L. Personal communication, V.A. Hospital, Houston, Texas, December 1987.

OCTOBER 1991 VOLUME 66 NUMBER 4

F L U O R I D E E F F E C T S ON I N T R A O R A L M A G N E T S

0.200"

• --*

SC

I - I

NI

0.200-

• --

• - - • oK

SC •

NI

. - - v

0,150-

0.150-

E

E c~°

*--*

0,100-

"-" ~:_______.

O. I O 0

~D

__

~

~ "

0~ 0.050-

0.000

@

@

SnF2

I

I

I

I

I

I

0

1

2

,3

4

5

SIMULATED EXPOSURETIME (YEARS)

o

NoF 0.000

I

I

I

I

I

I

0

I

2

3

4

5

SIMULATED EXPOSURE TIME (YEARS)

Fig. 1. Plots of mean Ra of four magnet types stored in SnF2. Plots include initial mean Ra as well as mean Ra for simulated 1, 2, and 5 years' exposure to fluoride solutions.

Fig. 2. Plots of mean Ra of four magnet types stored in NaF. Plots include initial m e a n Ra as well as mean Ra for simulated 1, 2, and 5 years' exposure to fluoride solutions.

the solution was changed, the pH was measured with a digital pH meter (Model 805 MP, Acumet, Fisher Scientific, Springfield, N.J.). After 31 hours (1 year of simulated exposure), the blocks containing the magnets were removed from the fluoride solutions, ultrasonically cleaned for 5 minutes in distilled water, dried with compressed air, and measured for Ra as described previously. This was repeated at 62 hours (2 years of simulated exposure) and at 155 hours (5 years of simulated exposure). The independent variables of this experiment are fluoride and magnet types. The dependent variable is Ra of the magnets (n = 7). As the initial m e a n Ra of the four magnet types were not the same, even after polishing SC and NI, the mean change in Ra from the initial value was used for statistical analysis and comparison. The mean change in Ra from the initial value is represented by 0-1, 0-2, and 0-5 for simulated 1, 2, and 5 years of exposure, respectively. Twoway analysis of variance (ANOVA) for 0-1, 0-2, and 0-5 years' simulated exposure was calculated to compare the factors of fluoride and magnet type. Individual mean changes in Ra due to fluoride and magnet type were compared using Tukey's HSD interval at the 95% confidence level. Tukey intervals were calculated at the 95% confidence level to compare mean change in Ra between magnet types, within fluoride types, and to compare mean change in Ra between fluoride types, within magnet types, all within the same period of simulated exposure. Data from repeated measurements on the same magnets (within fluoride) types for 0-1, 0-2, and 0-5 years' simulated exposure were compared using a paired t test.

2, and 5 years' exposure to the fluoride solutions. The mean change in Ra of the four magnet types stored in SnF2 and NaF are plotted in Figs. 3 and 4, respectively. The plots reflect the increase in m e a n Ra from the initial value due to simulated 1, 2, and 5 years' exposure to the fluoride solutions, labeled as 0-1, 0-2, and 0-5 years simulated exposure time, respectively. Two-way ANOVA of the mean change in Ra ascribed to fluoride and magnet type for 0-1, 0-2, and 0-5 years' simulated exposure is presented in Table II. The F ratios are all significant at p < 0.05, indicating that some difference exists between mean change in Ra because of fluoride or magnet type at each period of simulated exposure. Table III contains the mean change in Ra, standard deviations, and the Tukey intervals used to compare mean change in Ra between magnets within fluorides for each period of simulated exposure. Vertical lines indicate no significant difference (p < 0.05) between the mean change in Ra for magnet types connected, within the fluoride type indicated and within the same period of exposure. No difference was found in mean change in Ra between DM and DK after simulated 1, 2, or 5 years of exposure to either SnF2 or NaF. No difference was found in mean change in Ra between DM, DK, and SC after 1-year simulated exposure to either SnF2 or NaF. No difference was found after 2 years' simulated exposure to SnF2. The other significant differences are noted in Table III. After simulated 1, 2, and 5 years of exposure to either fluoride solution, magnet type NI had a greater mean change in Ra than the other magnets. Table IV contains the mean change in Ra, standard deviations, and the Tukey intervals used to compare mean change in Ra between fluorides within magnet types for each period of simulated exposure. Vertical lines indicate no significant difference (p < 0.05) between the mean change in Ra for fluoride types connected, within the mag-

RESULTS The mean Ra of the four magnet types stored in SnF2 and NaF are plotted in Figs. i and 2, respectively. The plots include initial mean Ra as well as mean Ra for simulated 1,

THE J O U R N A L OF P R O S T H E T I C D E N T I S T R Y

555

OBATAKEET AL

0.200-

0.200"

?

• -- • =-- • • --"

Sl

• -- •

DK

SnF2

OM

• --

*

• --"

o.~5o.

SC NI DU DK

NaF

0.150

/

0.100.

u~ 0 . 1 0 0

/

.E dj

II--ll

E

5

/

~"-

0.050

0.050

o,-, ~ 0.000

E

0-I

I

0.000

I

0-2

~

0-1

0-5

v ~

-

-

I

I

I

0-2

0-5

SIMULATED EXPOSURE TIME (YEARS)

SIMULATED EXPOSURE TIME (YEARS)

Fig. 3. Plots of mean change in Ra of four magnet types stored in SnF2. Plots reflect increase in mean Ra from initial value to 1, 2, and 5 years' simulated fluoride exposure-labeled 0-1, 0-2, and 0-5, respectively.

'

Fig. 4. Plots of mean change in Ra of four magnet types stored in NaF. Plots reflect increase in mean Ra from initial value to 1, 2, and 5 years' simulated fluoride exposure-labeled 0-1, 0-2, and 0-5, respectively.

Table III. Mean change in Ra* within fluoride types Year(s)

TI

Magnet

SnF2

Magnet

NaF

0-1

0.010

DK SC DM NI

0.003 (0.003) * 0.004 (0.004) 0.006 (0.005) 0.034 (0.011)

DM DK SC NI

0.005 (0.004) I * 0.005 (0.003) 0.012 (0.007) 0.031 (0.023)

0-2

0.011

DM DK SC N1

0.007 (0.005) 0.008 (0.003) 0.015 (0.007) 0.056 (0.012)

DM DK SC NI

0.007 (0.004) I 0.009 (0.004) I o.o25 (0.010) 0.072 (0.022)

0-5

0.012

DM DK SC NI

0.009 (0.005) 0.010 (0.004) 0.021 (0.009) 0.094 (0.024)

DM DK SC NI

0.013 (0.005) I 0.014 (0.005) I 0.029 (0.010) 0.112 (0.015)

n = 7. Standard deviations in parentheses. Vertical lines indicate no significant difference (p < 0.05). TI, Tukey interval at 95% confidence level. *Ra in microns.

net type indicated and within the same period of exposure. No difference was found in mean change in Ra for DM and DK because of fluoride type after simulated 1, 2, or 5 years' exposure. No difference was found in mean change in Ra for NI because of fluoride type after 1 year of simulated exposure. Significant differences were found because of fluoride type in mean change in Ra for SC after simulated 1, 2, and 5 years' exposure and for NI after 2 and 5 years' exposure. In each instance where differences were found, the mean change in Ra was greater with NaF than with SnF2 Table V presents the results of paired t tests for the mean change in Ra due to fluoride exposure. For each magnet type, in either SnF2 or NaF, the mean Ra after simulated 1, 2, and 5 years' exposure was significantly greater (p < 0.03) than the initial mean Ra. The p values are given

556

in Table V, along with the mean, standard error of the mean, and the t statistic.

DISCUSSION The development of SmCo5 magnets led to the widespread use of magnets in the retention of overdentures and in maxillofacial prosthetic applications. 6 The coercivity (retention of magnetism) of SmCo5 is five times that of CoPt and 10 times that of Alnico magnets, 7 thus permitting fabrication into small, thin shapes that retain high magnetic strength. The recently developed Nd-Fe-B magnets have slightly higher retentive qualities than SmCos. s Cerny9 and Tsutsui et al. 1° have shown that short-term exposure of tissues to SmCo5 magnetic fields has no adverse effects. The effects of long-term exposure of tissues to

OCTOBER 1991 VOLUME 66 NUMBER 4

FLUORIDE EFFECTS ON INTRAORAL MAGNETS

Table

IV. Mean change in Ra* between fluoride types Magnet type

Year(s)

TI

0-1

0.006

FL

SC

NI

DM

DK

0.006 (0.005) [ *

s

0.004 (0.004) *

0.034 (0.011)i *

N

0.012 (0.007)

0.031 (0.023)[

0.005 (0.004) [

0.003 (0.003) [ * 0.005 (0.003) [

I

0-2

0.006

S N

0.015 (0.007) 0.025 (0.010)

0.056 (0.012) 0.072 (0.022)

0.007 (0.005) I 0.007 (0.004) I

0.008 (0.003) ] 0.009 (0.004)

0-5

0.007

S N

0.021 (0.009) 0.029 (0.009)

0.094 (0.024) 0.112 (0.015)

0.009 (0.005) i 0.013 (0.005) ]

0.010 (0.004) I 0.014 (O.O05)

n = 7. Standard deviationsin parentheses. Vertical lines indicateno significantdifference (p < 0.05). FL, Fluoridetype (N = NaF; S = SnF2);other abbreviationsas in Table III. *Ra in microns.

Table V. Mean change in Ra*: Paired test--years of simulated exposure to fluoride Fluoride type SnF2 Magnet

Year(s)

Mean*

SEM

SC

0-1 0-2 0-5

0.004 0.015 0.021

0.001 0.003 0.003

NI

0-1 0-2 0-5

0.034 0.056 0.094

DM

0-1 0-2 0-5

DK

0-1 0-2 0-5

NaF T

p

Mean*

SEM

T

p

3.06 5.51 6.56

0.0220 0.0015 0.0006

0.013 0.025 0.029

0.003 0.004 0.004

4.75 6.83 8.09

0.0032 0.0005 0.0002

0.004 0.005 0.009

8.25 12:03 10.28

0.0002 0.0000 0.0000

0.031 0.072 0.112

0.009 0.008 0.006

3.49 8.46 19.81

0.0130 0.0001 0.0000

0.006 0.007 0.009

0.002 0.002 0.002

3.16 4.01 5.37

0.0200 0.0070 0.0017

0.005 0.007 0.013

0.002 0.002 0.002

2.85 4.49 7.24

0.0290 0.0041 0.0004

0.003 0.008 0.010

0.001 0.001 0.001

3.36 6.35 7.48

0.0150 0.0007 0.0003

0.005 0.009 0.014

0.001 0.001 0.002

3.65 6.22 6.83

0.0110 0.0008 0.0005

n~7.

*Ra in microns.

magnetic fields is unknown. Tsutsui et al. I° also recommended chromium plating SmCo5 magnets to reduce corrosion. The plated magnets were shown to have good corrosion resistance to artificial saliva, 0.1% Na2S, and to 1.0% NaC1 solutions. Gillingss recommended that stainless steel end plates be placed around the magnet to increase strength and corrosion resistance. Encapsulated intraoral magnets are encased in either PdCo alloys, stainless steel, acrylic resin, or a combination of these materials. Keepers are currently made from stainless steel or PdCo alloys. Kinouchi et al. 11 and Vrijhoef et al. 12 found that PdCo alloy keepers had ac-

THE JOURNALOF PROSTHETICDENTISTRY

ceptable corrosion-resistant properties to artificial saliva, Na2S, NaC1, lactic acid, and HC1 solutions. The results of this study indicate that the nonencapsulated Nd-Fe-B magnet (NI) corrodes more t h a n the other types tested at simulated 1, 2, and 5 years of exposure to SnF2 or NaF, using Ra as a criterion. The encapsulated magnet (DM) and its keeper (DK), both with a PdCo surface, demonstrated high resistance to corrosion in both fluoride solutions for simulated 1, 2, and 5 years' exposure. The nonencapsulated SmCo5 demonstrated a greater corrosion resistance to SnF2 t h a n NaF at simulated 1, 2, and 5 years' exposure.

557

OBATAKE ET AL

Magnet types SC and NI were polished to provide greater similarity in initial surface smoothness for the four magnets evaluated. Types SC and NI are not coated with protective material during manufacture, so polishing the surface should not affect the corrosiveness of these magnets-except possibly to decrease the corrosion by providing a smooth, scratch-free surface. Types DM and DK were not polished because the initial mean Ra was similar to that of types SC and NI after polishing. Mean initial Ra for all magnets were in the range of 0.040 to 0.133 ~m. Although differences exist in the mean change in Ra between magnet types and between fluoride types, the clinical significance of differences of this magnitude is unknown. This is particularly true when the many other factors involved in magnet corrosion are considered. However, corrosion is a concern with any metal used intraorally. Numerous variables affect the corrosion of magnet systems used intraorally. In addition to the fluorides tested in this study, other factors that may be expected to influence the rate and amount of corrosion of these magnet systems include: (1) saliva pH, composition, and consistency; (2) diet and nutritional status; and (3) masticatory habits. This study evaluated the effects of fluoride use on magnet systems tested in vitro, and the results may not necessarily reflect the rates or amounts that would be experienced in vivo. Corrosion is difficult to measure precisely. Methods typically used include electrochemical analysis and weight loss. The change in surface roughness is an indicator of the extent of corrosion that has occurred. However, the results are dependent on several qualifications. These include: (1) The size and force of application of the profilometer stylus tip, though extremely small, may have altered the roughness of the surface that was measured during the measurement operation because of the potentially fragile nature of the corroded surface. (2) Ultrasonic cleaning of the magnets to remove fluoride residue may have altered the corroded surface prior to surface roughness measurement. (3) Repeated measurements of the same magnet surface could introduce some error in roughness measurement. Future work should be directed toward correlating the surface roughness with electrochemical or weight loss measurements of corrosion. Using the surface roughness as an indicator of the extent of corrosion, it is possible to conduct clinical studies of intraoral magnets in service under a variety of conditions, since the surface roughness measurement is simple, quick, and nondestructive. Studies of this nature would permit accurate estimation of the lifeexpectancy of intraoral magnets before they require replacement. CONCLUSION

four types of intraoral magnets used for retention of overdentures and maxillofacial prostheses. Using Ra as an indication of corrosion, a comparison of the predicted im traoral corrosion of four intraoral magnet systems has been made. The results may not be directly applicable to intraoral corrosion because of intraoral variables not measured. Further studies will be necessary to correlate Ra with electrochemical corrosion and to evaluate the clinical factors that may affect intraoral corrosion. Based on this in vitro investigation, it appears that NaF rinses may cause more corrosion of nonencapsulated intraoral magnets used for retention of overdentures or maxillofacial prostheses than SnF2 rinses over a total exposure time of 155 hours. No significant difference was found in the change in Ra for PdCo encapsulated intraoral magnets because of fluoride type. All four magnet systems became rougher with continued exposure to either fluoride type, indicating continued corrosion. The magnet system that demonstrated the least change in Ra ascribable to exposure to fluoride was the PdCo encapsulated SmCo5 magnet and its PdCo keeper. REFERENCES 1. Miller PA, Complete dentures supported by natural teeth. J PROSTHET DENT 1958;8:924-8. 2. DerksonGD, MacEnteeMM. Effectof0.4% stannous fluoride gel on the gingival health of overdenture abutments. J PROSTHET DENT 1982; 48:23-6. 3. Toolson LB, Smith DE. A five-year longitudinal study of patients treated with overdentures. J PROSTHETDENT 1983;49:749-56. 4. Shannon IL, Cronin RJ. Clinical preventive dentistry for overdenture patients. In: Brewer AA, Morrow RM, eds. Overdentures. St Louis: CV Mosby Co, 1960:331. 5. Smith DC. Tissue reaction to noble and base metal alloys. Biocompatibility of dental materials, vol IV. Boca Raton: CRC Press, Inc, 1982:53. 6. Sasaki H, Kiuchi Y. Application of cobalt samarium alloy magnets in prosthetic restoration. Hotetsu Rinsho (Practice in Prosthodontics) 1976;9:77. 7. Gillings BRD. Magnetic retention for complete and partial overdentures. Part I. J PROSTHET DENT 1981;45:484~91. 8. Jackson TR, Healy KW. Rare earth magnetic attachments; the state of the art in removable prosthodontics. Quintessence Int 1987;18:41-51. 9. Cerny R. The reaction of denture tissues to magnetic fields. Aust Dent J 1980;255:264-8. 10. Tsutsui H, Kinouchi Y, Sasaki H, Shioto M, Ushita T. Studies on the Sm-Co magnet as a dental material. J Dent Res 1979;58: 1597~606. 11. Kinouchi Y, Ushita T, Tsutsui H, Yoshida Y, Sasaki H, Mizaki T. Pd-Co dental casting ferromagnetic alloys. J Dent Res 1981;60: 50-8. 12. Vrijhoef MMA, Mezger PR, Ven Der Zel JM, Greener EH. Corrosion of ferromagnetic alloys used for magnetic retention of overdentures. J Dent Res 1987;66:1456-9. Reprint requests to: DR. STEPHEN M. COLLARD DENTAL BRANCH UNIVERSITY OF TEXAS HEALTH SCIENCE CENTER 6516 JOHN FREEMANAVE. HOUSTON, WX 77030

This investigation has identified the in vitro effects of two common fluoride rinses on the change in surface Ra of

558

OCTOBER 1991 VOLUME 66 NUMBER 4