Enamel surface characteristics removal of bonded orthodontic
V. Burapavong, D.D.S., MS., G. W. Marshall, Ph.D., D.A. Apfel, D.M.D., M.S.D., and H. T. Perry, D.D.S., Ph.D. Chicago,
cid-etching techniques are now commonly used in many dental procedures for enhanced adhesion of resins to enamel. However, the only major technique using these procedures for temporary attachment is the direct bonding of orthodontic brackets. Therefore, this procedure requires debracketing at the termination of treatment and use of techniques which return the treated enamel surface to “normal.” Many studieslm4 have been conducted on the bond strength of resins to acid-etched enamel, but only recently have any serious attempts been made to critically evaluate proposed methods for removal of the adhesives after direct bonding.“* 6 Caspersen’ studied a single adhesive system with plastic brackets attached in situ and removed by four procedures, while Gwinnett and Gorelick studied three adhesives and a number of debonding procedures on extracted teeth. The present investigation used two different adhesives and a variety of removal and finishing procedures, applied in vivo, and sought quantitative and qualitative characterization of the enamel surfaces by scanning electron microscopy analysis. Materials and methods Thirty-six maxillary or mandibular first or second premolars which were scheduled for extraction for orthodontic purposes, in twelve volunteer patients, were bonded with metal brackets* using either the NT or G$ bonding system. Three additional premolars with untreated surfaces were used as controls. Prior to acid etching, the teeth were polished with fluoride-free pumice, washed, and dried with oil-free air. Half of the experimental teeth were bracketed with each adhesive, following each manufacturer’s directions explicitly for etching, mixing, and application procedures. The brackets were left in place for one week to allow for complete polymerization and near maximum bond strengths of the adhesives. *, 5 The brackets were removed with a From the Departments of Biological Materials and Orthodontics, Northwestern University Dental School. This work was supported in part by Transon Corp., Fenton, Missouri, National Institutes of Health Grant 5507.RR053 11, and National Institute of Dental Research Grant DE 04000. Based on a thesis presented by the senior author for the M.S. degree, Northwestern University Dental School, 1977. *0.08 inch slot metal edgewise bracket, GAC Int., Inc., Farmingdale, N. Y. tNuva-SeaI/Nuva-Tach, UV activated adhesive, L.D. Caulk Co., Milford, Del. #Genie cold-curing adhesive. Lee Pharmaceuticals, S. El Monte, Calif. 0002-9416/78/0274-0176$01.20/0
1978 The C. V. Mosby CO.
Volume 74 Number 2
G G G G G G N N N N N N Controls
Hand scaler Green stone Ultrasonic scaler Hand scaler Green Stone Ultrasonic scaler Hand scaler Green stone Ultrasonic scaler Hand scaler Green stone Ultrasonic scaler
None None None Pumiced Pumiced Pumiced None None None Pumiced Pumiced Pumiced
No. of teeth
3 3 3 3 3 3 3 3 3 3 3 3 3
ligature wire cutter* and the residual adhesive was removed with either a hand scaler,?’ green stone$ at low speed, or an ultrasonic scaler. QEach technique was assigned randomly in each of the four quadrants for each patient. Thus, for a given patient, the same materials were applied and only the method of adhesive removal differed. Half of the samples received a final pumicing treatment and half were left unpumiced. This gave a total of twelve experimental groups, as shown in Table I. The teeth were extracted the same day as bracket removal and stored in SOpercent ethanol, following Boyde’s7 suggestion, until preparation for study with the scanning electron microscope (SEM). The teeth were mounted on stubs and sputter-coated/l with a thin layer (-2OOA”) of gold for conductivity prior to SEM examinations at 18 kv. beam potential. Micrographs of the debonded area were obtained at various magnifications from x 25 to x 1,000 at 0” tilt (tooth surface perpendicular to the beam) for qualitative assessment of each surface and to estimate areas covered with adhesive and roughened by the removal treatment. Stereopair photomicrographs were obtained at -5 and +5” tilt for parallax measurements to determine depth of damage to the enamel. The area of retained adhesive was determined by means of the point counting method for quantitative metallography.* In this method, a 1 by 1 inch grid of 100 points was applied randomly ten times to each area of interest. The area1 fraction, A, of retained adhesive is then obtained from”:
where n4 15the number of points foliowing within adhesive and n is the total number of points ( 1,000 per analysis). This equation has an associated statistical variance of +(A.,)
*ETM, Monrovia, CaIif. tT,/T, Hu-Friedy Mfg. Cu., Chicago. 111. $The Torit Corp., St, Paul, Mimi. PTransonic UD-65, Transon Corp., Fenton, MO. IlHummer II. sputter-etching unit, Technics, Inc., Alexandria, Va. IISI Super Mini SEM, International Scientific Instruments, Mountain
Am. J. Orrhod. August 1978
where AA is the area1 fraction of the smaller component. The area1 fractions were convereted to percentages for all samples. Measurements of the maximum depth of damage were made by viewing stereopairs of each sample in a stereoviewer* and choosing the area of maximum damage. Parallax measurements were made between corresponding points on the two images and were used to determine the vertical separation between the top and bottom of the hole from”: MZ
where Z is the height difference, M is the magnification, P is the parallax, and 8 is the angle of tilt between the two micrographs. In all cases, the depth measurements were corrected for local inclination of the tooth from a second parallax measurement. An additional measurement of the damaged enamel was sought through measurements of an area of roughness. The point-counting method described previously was applied to the areas undergoing treatment which were detectably roughened. The minimum area considered to be roughened was estimated from stereopairs to be 5 pm in depth. This gave a definition of the area of roughness as that portion of the treated surface which exhibited damage to a depth equal to or exceeding 5 pm. Results Initially, the nature of the enamel surfaces was qualitatively compared using the SEM. Although distinct differences could be seen for each removal procedure, the surfaces bracketed with either N or G were similar for a given removal technique. Fig. 1 compares the surfaces left after the hand scaler treatment for each material. Occasional islands of adhesive (A) were left, along with a large number of shallow scratches. Abrasive striations were left after the green stone treatment for both materials, as shown in Fig. 2 for system N. The ultrasonic scaler left surfaces with many small gouges and depressions. A final pumicing treatment left most surfaces virtually free of retained adhesive and considerably smoother than the unpumiced surfaces, as illustrated in Fig. 3. Fig. 4, at higher magnification, shows the enamel with the surface prepared by hand scaler and pumicing and demonstrates restoration of the surface to near “normal.” However, the surfaces treated with the green stone followed by pumice always left evidence of the abrasion striations, as shown in Fig. 5. The pumiced surface prepared by the ultrasonic scaler was generally as smooth as the hand scaler treatment. All the treatments left occasional gouges in the enamel which were smoothed, but not removed, by final pumicing, as illustrated in Fig. 6 for the ultrasonic scaler. Thus, qualitatively, there was little difference in behavior of the adhesive systems and all treatments left adhesive on the surfaces which could be removed by pumicing. All treatments also left occasional scratches, depressions, and a few gouges on the surfaces. The majority of the defects from the hand scaler and ultrasonic scaler were removed by pumicing, but the cutting lines from the green stone were not. Pumicing did not remove the deeper gouges left by any treatment. The results of the point count analysis for the area1 percentage of retained adhesives *Folding
Volume 74 Number 2
Fig. 1. SEM photomicrographs Systen n G. 8, System N.
Enamel surface and bonded brackets
tooth surfaces E, Enamel.
seal er. A,
are co nveniently illustrated in the bar graph of Fig. 7. Only those differences which were signifi scantly different in a t test of the means at P < 0.05 will be noted. All pretreatn nents (hand scaler, green stone, ultrasonic scaler) left considerable amounts of adhesive 01n the teeth. The hand scaler pre-treatment left more adhesive than the other treatments wh en G was u sed. However, large amounts of adhesive were generally left when N was used and,
Am. J. Orrhod. AU&YLS~1978
Fig. 2. SEM photomicrographs of tooth striations covered most of the surface.
N with green
therefore, significantly more was left as compared to G when either the green stone or the ultrasonic scaler was used as the pretreatment. In all cases, this residual adhesive was removed on pumicing and left no difference, compared to the controls. The evaluation of enamel damage due to bracket removal was carried out in two stages. As an initial measure, the maximum depth of holes on each surface was determined from the stereopairs. This procedure provided little information which discriminated one technique from the other. Maximum damage was up to 20 /.L, and when all treatments were pooled G had a I5 pm average while N had about 9 pm. Both were significantly greater than the control (average, 2 pm) and were different from each other only in the pooled data. Thus, major differences in materials and techniques were not clearly indicated by the maximum damage measurement (see discussion). The area detectably roughened by each technique was determined in order to provide a better guide to enamel damage. The results of this evaluation are presented in Fig. 8. Again, only statistically significant differences (P < 0.05) will be noted. For both adhesive materials, the green stone gave the largest roughened area. For adhesive G, the other treatments also gave enamel surfaces rougher than the control. However, teeth given final pumicing treatments yielded reduced areas of roughness and no statistically significant differences for technique or material were found, compared to the control. Discussion
The intention of this study was to evaluate several techniques for cleaning the tooth surfaces after removal of directly bonded brackets. The qualitative SEM evaluation showed no striking differences between the two resin systems (N,G) employed in the
Volume 74 Number 2
Fig , 3. Pk
Enamel surface and bonded brackets
trea.tmt ,nt bonding procedure. After the initial removal of adhesive, with either a green stone, hand scaler, or ultrasonic scaler, the enamel surface retained some adhesive, as was also found in Casperson’s’ investigation. However, this residual adhesive was largely removed by pumicing, in contrast with Casperson’s results. It should be noted that the detection of residual adhesive may be difficult when the adhesive is in a thin film. If the islands of residual adhesive are large, they are easy to pick
Am. 1. Orthod. August 1978
Fig. 4. tiigher-magnification photomicrograph sive ren nova1 by hand scaler.
Volume 74 Number 2
Enamel surface and ban&d
Fig. 5. SEM photomicrograph stone and pumice treatments.
6. I dicrograph
by n neai
showing typical ultrasonic scaler and
of adhe sive
Am. J. Orfhod. August 1978
PUMICED cl CONTROLS = 0%
ADHESIVE Fig. 7. Bar graphs treatments shown.
I , I : I
demonstrating the percentage of tooth surface adhesives. I bar = + one standard deviation,
with retained controls were
adhesive for the various 0 percent and are not
out, but dark striations could be either a thin film of adhesive or damaged enamel. The secondary electron output from the adhesive should be reduced as compared to enamel, giving it the darker contrast. However, experiments in our laboratory have shown that dragging a hand scaler across etched enamel gives similar contrast. Casperson” used energy-dispersive x-ray spectroscopy (EDS) to aid in the adhesive discrimination. However, where gold coatings are used, this procedure has not been completely successful in our laboratory. This is largely the result of inconsistency in the Ca/Au and P/Au x-ray peak intensity ratios from etched enamel compared to enamel covered with resin in thin film. Where the resin film is thick, there is a predictable decrease in Ca and P intensities, but in thin films Au excitation and absorption often mask the desired information, particularly on etched enamel surfaces where it is difficult to correct for the geometry of the porous surface. This suggests that the detection of thin films of resin can be difficult, with or without EDS, and is a source of error in both the qualitative and quantitative assessments. The presence or absence of such thin films may not be of clinical importance, since Silverstonel” suggested that such thin resin films do not increase enamel solubility. Furthermore, any color change in the film probably could not be perceived. Qualitatively, the roughness of the tooth surfaces was reduced by the final pumicing treatment, at least to some degree, as pointed out in previous reports.“, 63 I1 The hand scaler pretreatment left scratches which could be removed by pumicing, although gouges occasionally occurred because of the hard instruments, as noted by Gwinnett and Gorelick.‘j The green stone removal procedure left definite cuts and striations in the enamel surface, some of which remained after pumicing, as also shown by Gwinnett and Gorelick. The ultrasonic scaler left numerous small depressions in the surface which could be generally smoothed by pumicing and, therefore, this appeared to be a viable technique for adhesive removal. Current work being conducted in our laboratory indicates that this finding cannot be generalized for all ultrasonic instruments or tips and the result is specific for the instrument used in this investigation. The measurements of retained adhesive (Fig. 7) showed that all of the techniques left
Volume 74 Number ?
I I I ;
Fig. 8. Bar graphs demonstrating the percentage of detectably ments and adhesives. I bar = 2 one standard deviation.
appreciable amounts of adhesive on the surface. For system G, the hand scaler left the most (57 percent), while the ultrasonic and green stone left 24 percent and 20 percent, respectively. For adhesive N, all three techniques left more than 50 percent of the area covered with adhesive. This difference in behavior between the two systems could be related to several factors, including the fact that system N is a two-component system with the initial layer being a low-viscosity resin which easily penetrates the etched enamel. Thus, much of the retained adhesive for N appears in thin film form which clings tenaciously to the enamel surface and might be more easily separated from the filled component than from the enamel. Such observations are supported by the study by Brose and Associates’” in which it was found that remnants of the low-viscosity resin component of composites were often found on fracture surfaces. This feature was attributed to interpenetration of the enamel by the resin. Second, the thin film measurements are subject to greater error, as previously described, and this might contribute to the slight differences in the two systems. Virtually all the adhesive was removed for both resins and all pretreatments after a final pumicing treatment. It should be noted that if the film was very thin it would not appear in these measurements and any resin which had interpenetrated the enamel to form tags would not be detected. Within these limitations, all the adhesive was removed and pumicing was a necessary and effective final step in adhesive removal. Perhaps the most important factor to consider in a bracket-removal technique is the damage done to the enamel surface. Two measures of damage were considered, namely, the maximum depth and the area roughened. The maximum depth ranged from about 10 to 20 pm and was largely independent of removal technique, adhesive, or pumicing. This suggests that any of the three techniques can introduce unwanted damage to the enamel surface. but. since the measurement was made on a single point of each surface, it was not indicative of total enamel damage and could not discriminate between procedures. Until superior cleaning procedures are developed, occasional holes in the enamel, on the order of 10 to 20 pm deep, will apparently be a consequence of the debracketing procedure.
Am. J. Orthod. AU@JSI 1978
There was a slight difference in adhesive systems, with G giving slightly deeper gouges on the average than N. This could be a result of the differences in the two-component technique compared to the single-component G system, or to differences in bond strength of the two systems.’ However, the clinical significance of such small difference is uncertain. An improved measure of enamel damage was sought in the area detectably roughened. It is clear from Fig. 8 that the treatment by green stone leaves the area much rougher than any other treatment for both resins. This results from the fact that, on application, the green stone cut into the enamel before the operator thought the surface was clean. Thus, it would appear that the green stone should not be used in close proximity to the enamel surface. Much of this roughness was reduced by the final pumicing treatment, so that the surface treated by the green stone followed by pumicing was not significantly rougher than the others. For the initial treatments by ultrasonic or hand scalers, there were no significant differences in roughness for either technique. For system N, there was no increased roughness with these techniques as compared to the controls, while for system G the roughness was slightly but significantly greater. These small differences may be related to the variations in these systems noted earlier. It is important to note that after pumicing there was no significant difference in the roughness for any system or technique as compared to the control. Although pumice is a mild abrasive, it apparently is capable of smoothing most of the roughened areas left by the other techniques. Only the deep gouges are not removed by the pumice treatment. Thus, it appears that final pumicing serves effectively to remove residual adhesive and restore smoothness to the treated enamel surface. It is further suggested that both hand scalers and ultrasonic scalers are effective for initial adhesive removal. The green stone is unnecessarily damaging to the enamel surface, and its use should probably be limited to areas well away from the enamel surface. Summary
Thirty-six premolars which were scheduled for routine extraction were bracketed with either N or G system adhesives. After one week the brackets were removed and the residual adhesive removed by low-speed green stone, a hand scaler, or an ultrasonic scaler. Half the teeth were given a final pumice, and then all were extracted. The SEM was used to evaluate the enamel surface, and quantitative assessments were made for the area covered with retained adhesive and for damage to the surface. All three removal techniques left appreciable amounts of retained adhesive 20 to 60 per cent coverage for G and 55 to 75 per cent for N adhesive. All detectable adhesive was removed by final pumicing. The detectably roughened area was over 50 per cent for the green stone treatment for both materials and appeared to introduce unnecessary roughness. The hand scaler and the particular ultrasonic scaler used in this study did not severely roughen the enamel for either material and appeared to be effective for initial adhesive removal. All three techniques introduced occasional gouges 10 to 20 pm in depth which could not be removed. All other roughness was generally smoothed by final pumicing. Thus, pumicing is a necessary final step with all removal procedures studied. REFERENCES I. Moser, J. B., Dowling, D. B., Greener, E. H., and Marshall, , . _ _ -- ,.. I,.--e I... . .__., _. yc‘I._ .\iJ. JJ. -r, I-t,o, 1‘1,“.
G. W.: Adhesion
Volume 74 Number 2
Enamel surfuce and bonded brackets
2. Green, F. P.: An in vitro study of five direct bonding orthodontic systems utilizing polycarbonate brackets, M.S. thesis, Northwestern University Dental School. Chicago, 1976. 3. Retief, D. H.: Failure at the dental adhesive-etched enamel interface, J. Oral Rehab. 1: 265-284, 1974. 4. Wickwire, N. A., and Rentz. D.: Enamel pre-treatment: A critical variable in direct bonding systems, AM. J. ORTHOD.
5. Casperson, microscope 6. Gwinnett, AM.
I.: Residual acrylic adhesive after removal of plastic orthodontic brackets: study, AM. J. ORTHOD. 71: 637-650, 1977. A. J., and Gorelick, L.: Microscopic evaluation of enamel after debonding:
A scanning Clinical
7. Boyde, A.: Enamel structure and cavity margins, Oper. Dent. 1: 13.28, 1976. 8. Hilliard, J. E.: Measurement of volume in volume. In Dehoff, R. T., and Rhines, F. H. (editors): Quantitative microscopy, New York, 1968, McGraw-Hill Book Company, Inc. 9. Howell, P. G. T.. and Boyde, A.: Comparison of various methods for reducing measurements from stereopair scanning electron micrographs to real 3-D data. In Scanning electron microscopy/ 1972, Chicago, 1972, IIT Research Institute, pp. 234.240. IO. Silverstone, L. M.: Fissure sealants: The susceptibility to dissolution of acid etched and subsequently abraded enamel in vitro, Caries Res. 11: 46-5 I, 1977. I I. Norman, G. V., and Facq, J. M.: The effects of adhesive systems on tooth surfaces, AM. J. ORTHOD. 59: 67-75,
12. Brose, J., Moser, J. B., Marshall, incisors, Ill. Dent. J. 45: 273-277,
G. W., 1976.