Comparison of three standards for quantifying occlusal loss of composite restorations R.W. Bryant Department of Clinical Dentistry Westmead Hospital Dental Clinical School Westmead, NSW, 2145, Australia Received April 26, 1989 Accepted October 25, 1989 Based on a paper presented at the 18th Annual Session of the American Association for Dental Research, San Francisco, 1989 Dent Mater 6:60-62, January, 1990

Abstract-Occlusal loss of resin composite in posterior restorations may be measured by assessing the exposure of cavity walls. The purpose of this study was to compare three standard scales used to quantify occlusal loss on stone replicas of posterior composites. Measurements of occlusal loss, on 60 stone replicas selected from two clinical studies, were obtained by comparison with three standard scales: L-calibrated die stone standards derived from clinical restorations (Leinfelder), M-dies with cylindrical incremental defects (M-L scale), and V-tooth-sized dies with restoration-like incremental defects (Vivadent). Evaluatorswith different dental and evaluating experience carried out the assessment using each standard on two different occasions. Differences among results obtained using the three standard scales were comparatively small when assessments were carried out by experienced evaluators. Use of the V standard consistently achieved a higher level of agreement by pairs of evaluators.

he assessment of the amount of occlusal loss or wear of composite restorations in posterior teeth has increasingly made use of indirect (non-clinical) evaluation techniques that are able to quantify the wear (Leinfelder, 1987) rather than direct (clinical) techniques based on USPHS criteria (Cvar and Ryge, 1971). Modifications to resin composites and improvements in their wear characteristics have resulted in the need for more discriminating techniques of evaluation to detect smaller clinical changes. Indirect evaluation techniques, that have categorized the occlusal loss on cast replicas according to a set of standards, have been shown conclusively to be more sensitive in detecting early wear (Goldberg et al., 1984; Leinfelder et al., 1986). They also provide a means of identifying small increases in occlusal loss over a period of time (Goldberg et al., 1984; Vann et al., 1988). Other methods for assessing wear have included volume loss (Dennison et aI., 1980; Vrijhoef et al., 1985). and rank ordering techniques (Goldberg et al., 1984; Leinfelder et al., 1986). Two types of standards have been used in i n d i r e c t e v a l u a t i o n s to quantify the occlusal loss on cast replicas of posterior composite restorations. Goldberg et al. (1984) developed a set of four standard casts poured from impressions of clinical restorations. The modification of this system by Leinfelder et al. (1983), incorporating a standard set of five die stone casts, has been widely used in the quantitative assessment of occlusal loss. A second type of standard scale was developed by Moffa and Lugassy (1986) and consisted of cylindrical defects at visually perceptible intervals. A modification of this type of standard scale was recently developed by Rheinberger for Ivoclar/Vivadent (Schaan, Liechtenstein).

T

60 BRYANT~THREE STANDARDS FOR QUANTIFYING OCCLUSAL LOSS OF COMPOSITES

The aims of this study were to evaluate three standard scales used to quantify the occlusal loss on cast replicas of posterior resin composite restorations and to identify some evaluator factors that might influence the assessment. MATERIALS AND METHODS

Three standard scales were used to assess the occlusal loss or wear of posterior composite restorations. The standard scale developed by Leinfelder (Code L) consisted of five calibrated standard casts of permanent posterior teeth, restored with resin composite and exhibiting approximately 100 ~m to 500 ~m of occlusal loss in i n c r e m e n t s of approximately 100 ~m. Actual values for the average wear on each of these standards have been reported by Vann et al. (1988), and these values have been used in subsequent calculations. The M-L Scale (Code M), developed by Moffa and Lugassy (1986), contained 18 dies with cylindrical incremental defects ranging from 0 ~m to 1000 ~m. Vivadent provided an 18-category standard scale (Code V) that consisted of toothshaped, tooth-sized dies with restoration-like incremental defects ranging from 25 ~m to 1000 ~m. Unlike the defects of irregular depth on L standards, the extent of the "loss" on M and V standards was even around the entire margins of the restorations. Stone replicas were obtained from impressions of composite restorations in the posterior teeth of patients. F r o m replicas of several hundred restorations, a test set of 60 replicas was selected to provide examples of a relatively wide range of wear from zero to 300 txm on the occlusal margin of the worse side (buccal or lingual). Four evaluators (Codes R,C,T,J) had different levels of previous dental and evaluating experience (Table

1). Regardless of experience, all evaluators undertook a training program. The M-L calibrating set was used with each standard scale until an 85% level of accuracy was achieved (Moffa and Lugassy, 1986). On each occasion, immediately before evaluation of the test set of replicas, the M-L calibrating set of casts was used to recalibrate the evaluator against the respective standard scale to be used in the evaluation. The set of 60 test replicas was assessed six times by each evaluator (using each of the standard scales twice) over a period of three weeks. Evaluations of the test replicas were carried out in a random order by visual inspection at a magnification of 2X, using a point source of illumination directed so t h a t the light tended to cast shadows on the exposed cavity walls. Occlusal loss, or wear, on the worse side (buccal or lingual) of the occlusal margin of each test replica was determined by evaluation of the extent to which this axially-directed wall of the original cavity preparation was exposed in the test replica compared with the mentally averaged extent of occlusal loss around the entire margins of the standards. Occlusal loss represented the distance from the cavity margin to the surface of the composite restoration (Vann et al., 1988). Because the amount of occlusal loss was usually uneven around the margin of the test replica, evaluators were asked to select the standard or the interval between two standards that most closely represented the amount of occlusal loss on the test replica. If the occlusal loss was less than or equal to 200 ~m, assessment was made to the nearest 25 ~m. If wear was considered to be greater than 200 ~m, the assessment was made to the nearest 50 ~m. These assessment specifications reflect the capacity of evaluators to identify, visually, smaller differences when the actual amount of existing loss is less. The two m a n u f a c t u r e d s t a n d a r d scales reflected this, the M standard having 50-~m increments for defects greater than 100 ~m and the V standard scale having 50-~m increments for defects greater than 200 ~m but 25-~m increments below these levels.

TABLE1 THE EVALUATORS

Code

R C T J

Dentally Qualified Yes Yes Yes No

Photographs of Amalgams Yes Yes No No

TABLE2 INTRA-EVALUATORAGREEMENT

Percentage Agreementto within one Category* Between Repeated Evalu~ions M V L Evaluator Standard Standard Standard R 97 90 90 C 98 93 87 T 60 82 72 J 60 65 78 *One category = 25 ~m if wear _< 200 #m. One category = 50 ~m if wear > 200 #m.

For assessment of the differences between repeated measurements, between two standards, or between two evaluators, paired comparisons were made using a t test to test Ho:~D = 0 (where D is the difference) at an alpha level of 0.01 (Snedecor and Cochran, 1980). RESULTS AND DISCUSSION Table 2 shows for each evaluator and each standard the percentage of the 60 repeated assessments for which there was agreement to within one category, where one category was defined as 25 jxm or 50 ~m according to whether the amount of wear was assessed as not exceeding 200 ~m or greater than 200 ~m, respectively. No intra-evaluator difference was statistically significant (~ = 0.01). Two evaluators (R,C), with previous experience of "mental avera g i n g " in i n d i r e c t evaluations, achieved a high level of intra-evaluator agreement to within one category regardless of the standard scale. The two evaluators without previous experience in indirect evaluations (except for the t r a i n i n g program) had a much higher level of d i s a g r e e m e n t between r e p e a t e d evaluations for most standards. For each evaluator, the percentage agreement to within one category of results from two standards

Previous Experience in Indirect Evaluation Techniques Stone Replicas Stone Replicas of Amalgams of Composites Yes Yes No No No No No No

TABLE3 PERCENTAGEAGREEMENTFOR PAIRS OF STANDARDS FOR EACHEVALUATOR PercentageAgreementto within One Category* of Results from Two Standards M and V M and L V and L Evaluator Standards Standards Standards R 95 88 83 C 92 85 80 T 60 50 47 J 48 40 48 *One category = 25 l~m if wear _< 200 #m. One category = 50 p,m if wear > 200 p,m.

is summarized in Table 3. The data emphasize differences between experienced evaluators and those inexperienced in indirect evaluations. The percentage agreement between results for two standards for experienced evaluators was between 80 and 95%. Inexperienced evaluators commonly recorded less than a 50% agreement. Statistically, values obtained when the M standard or the L standard was used were significantly greater (a = 0.01) than values recorded when the V standard was used for all evaluators except evaluator R, for whom there was no significant trend when results for standards M and V were compared. Although the percentage level of agreement to within one category differed among evaluators, no evaluator exhibited a statistically significant trend when results for M and L standards were compared. Table 4 shows the average occlusal loss assessed by each evaluator for the 60 replicas using each standard. Because the replicas were intentionally selected to cover a wide range of wear, values for one standard deviation were large (from 28 ~m to 89 ~m). It is evident, however, that the V standard achieved the most uniform average values among the four evaluators.

Dental Materials/Janua~w 1990 61

TABLE4 AVERAGEOCCLUSALLOSS FOR 60 REPLICAS

Average Loss (l~m) for 60 Replicas M V L Evaluator Standard Standard Standard R 81 (45)* 77 (48) 92 (61) C 94 (30) 85 (28) 93 (45) T 119 (52) 92 (42) 133 (89) J 132 (53) 99 (45) 140 (77) *Value for one standard deviation in parentheses. TABLE5 PERCENTAGEAGREEMENTFOR PAIRSOF EVALUATORSFOR EACHSTANDARD

Percentage Agreement to Within one Category* of Results from Pairs of Evaluators** Standard R,C R,T R,J C,T C,J T,J M 73 48 42 53 52 53 V 77 66 73 70 78 70 L 70 37 47 47 48 45 *One category = 25 ~m if wear _< 200 r~m. One category = 50 i~m if wear > 200 ~m. **Evaluators are coded R, C, T, and J. TABLE6 SUBJECTIVEASSESSMENTOF STANDARDS BY EVALUATORS

Subjective Assessment of Standards (M, V, L) Standard Providing Slowest or Most Most Confidence Fastest Difficult aboutthe Evaluator Evaluation Evaluation Result R L V V C L M V T M L V J V L V

For each standard, the percentage agreement to within one categ o r y of t h e r e s u l t s for p a i r s of evaluators is shown in Table 5. The best levels of agreement were 73% for M standard, 78% for V standard, and 70% for L scale and were generally achieved by the two evaluat o r s e x p e r i e n c e d in i n d i r e c t evaluations. In contrast to standards M and L, the V standard achieved a comparatively good level of agreement for all pairs of evaluators. After completing the study, the e v a l u a t o r s w e r e asked independently to record subjective assessments of the standards (Table 6). While the evaluators who were ex-

perienced in indirect evaluations considered that the use of the L standard provided the fastest evaluation, the two inexperienced evaluators considered the L standard as either the most difficult or slowest to use. All four evaluators, particularly the inexperienced evaluators, indicated that they were most confident of their assessments when they used the V standard. The findings consistently indicated that evaluators who were experienced in the "mental averaging" required for indirect evaluations achieved higher levels of intra-evaluator, inter-evaluator, and i n t e r standard agreement. Previous dental experience did not appear to be as important a factor. It may be nec' essary for evaluators with no previous indirect evaluation experience to undergo an extended training program in which they assess a set of trial replicas exhibiting a known or well-accepted amount of occlusal loss. This study examined precision and compared results obtained when different standards were used. It did not specifically examine accuracy which, in the evaluation of the wear or occlusal loss on posterior composite restorations, is assumed to have been achieved when there is a high level of intra-evaluator and interevaluator agreement among experienced evaluators. CONCLUSIONS

The previous experience of the evaluator in indirect evaluations is of greater importance than the choice of standard scale by which to quantify the occlusal loss on posterior composite restorations. Inexperienced evaluators, who have no previous experience in "mental averaging" for indirect evaluations, may require a substantial training program. Differences among r e s u l t s obtained by use of t h r e e s t a n d a r d scales, that were designed to assist in quantification of occlusal loss, were comparatively small when assessments were carried out by experienced evaluators. The use of a standard, consisting of tooth-shaped, tooth-sized dies with regular, restoration-like, incremen-

tal defects, (i) provided the most uniform set of average wear values among the four evaluators; (ii) consistently achieved a higher level of agreement by pairs of evatuators; and (iii) provided evaluators with the greatest sense of confidence in their assessments.

ACKNOWLEDGMENTS The author would like to thank Dr. J.C. Mitchem and Dr. D.B. Mahler for their suggestions and advice and Dr. C.J. Collins, Dr. T.M: Gerzina, and Mrs. J. Bell for their assistance with the evaluation. REFERENCES

CVAR, J.F. and RYGE, G. (1971): Criteria for the Clinical Evaluation of Dental Restorative Materials. USPHS Pub. No. 790-244, San Francisco: US Government Printing Office. DENNISON, J.B.; POWERS, J.M.; and CHARBENEAU, G.T. (1980): Measurement of in vivo Wear on Posterior Composite Restorations, J Dent Res 59: 318, Abst. No. 202. GOLDBERG, A.J.; RYDINOE, E.; SANTUCCI, E.A.; and RACZ, W.B. (1984): Clinical Evaluation Methods for Posterior Composite Restorations, J Dent Res 63: 1387-1391. LEINFELDER,K.L. (1987): Evaluation of Criteria Used for Assessing the Clinical Performance of Composite Resins in Posterior Teeth, Quint Int 18: 531536. LEINFELDER, K.L.; BARKMEIER,W.W,; and GOLDBERG, A.J. (1983): Quantitative Wear Measurements of Posterior Composite Resins, J Dent Res 62: 671, Abst. No. 194. LEINFELDER, K.F.; TAYLOR, D.F.; BARKMEIER, W.W.; and GOLDBERG, A.J. (1986): Quantitative Wear Measurement of Posterior Composite Resins, Dent Mater 2: 198-201. MOFFA, J.P. and LuGAssY, A.A. (1986): Calibration of Evaluators Utilizing the M-L Occlusal Loss Scale, J Dent Res 65: 302, Abst. No. 1197. SNEDECOR, G.W. and COCHRAN,W.G. (1980): Statistical Methods, 7th ed. Ames, Iowa: Iowa State University Press. VAN~, W.F., Jr.; BARg.MEZER, W.W.; and ]~AHLER, D.B. (1988): Assessing Composite Resin Wear in Primary Molars: Four-year Findings, J Dent Res 67: 876-879. VRIJHOEF, M.M.A.; HENDRIKS, F.H.J.; and LETZEL, H. (1985): LOSS of Substance of Dental Composite Restorations, Dent Mater 1: 101-105.

62 BRYANT~THREE STANDARDS FOR QUANTIFYING OCCLUSAL LOSS OF COMPOSITES

Comparison of three standards for quantifying occlusal loss of composite restorations.

Occlusal loss of resin composite in posterior restorations may be measured by assessing the exposure of cavity walls. The purpose of this study was to...
378KB Sizes 0 Downloads 0 Views