R

ABSTRACT
Background. The purpose of the authors in vivo investigation was to analyze the characteristics of noncarious cervical lesions, or NCCLs, in adult patients who had a high incidence of them. Methods. The patient pool consisted of a total of 57 patients and 171 teeth (three teeth per patient), with one NCCL per tooth. The characteristics the authors evaluated were shape, dimensions, sensitivity, sclerosis and occlusion. Results. In terms of lesion characteristics, 91 percent of the lesions had axial depths of 1 to 2 millimeters, 49 percent had occlusogingival widths of 1 to 2 mm, 74 percent had an angular shape of 45 to 135 degrees, 76 percent had mild or moderate sclerosis, and 73 percent had no or mild sensitivity. In terms of occlusion, 75 percent of teeth had an Angle Class I occlusion on the involved side, 60 percent had group function or mixed excursive guidance, 82 percent had wear facets, and 99 percent had Type 0 or I mobility. In terms of tooth location, 70 percent of NCCLs were on posterior teeth, 65 percent were on maxillary teeth, and 46 percent were on premolars. Conclusions. The evaluated NCCLs were found mainly to have small dimensions of depth and width (< 2 mm) and to be roughly right-angled in shape, and many had sclerosis and low sensitivity. A majority of the dentitions studied had Class I occlusion, with group function, prevalent wear facets, and little or no mobility. Cervical lesions were more common with posterior maxillary teeth and premolars, especially first premolars, which had the highest prevalence of lesions. Older patients were more likely to exhibit noncarious cervical lesions, but no great difference in incidence was found between men and women. Clinical Implications. A knowledge of the NCCL characteristics and etiologic covariables aids in proper case selection for treatment, aids in selection of appropriate treatment protocols and improves assessment of prognosis.

Characteristics of noncarious cervical lesions

A clinical investigation
TAR C. AW, D.D.S., M.S.; XAVIER LEPE, D.D.S., M.S.; GLEN H. JOHNSON, D.D.S., M.S.; LLOYD MANCL, Ph.D.

noncarious cervical lesion, or NCCL, is the loss of tooth structure at the cementoenamel junction, or CEJ, level that is unrelated to dental caries. These lesions can affect tooth sensitivity, plaque retention, caries incidence, structural integrity and pulpal vitality.1-6 The NCCL is being seen with increasing frequency and presents unique challenges for successful restoration.4,7,8 Background and review of literature. The prevalence of cervical lesions has been reported to be from 5 to 85 percent in various study populations.1,4,7,9-13 To properly treat such a lesion, it is important to consider its etiology. The CEJ is Older patients an area of structural weakness where were more the enamel layer is at its thinnest.14 likely to exhibit Erosion, abrasion and abfraction (stress noncarious flexure) are believed to be causative in in this vulnercervical lesions, the formation of NCCLs able area of enamel.15,16 Erosion is the but no great chemical dissolution of tooth structure difference in by acids, which can be intrinsic or incidence was extrinsic in origin.17 Abrasion is the found between mechanical wear of tooth structure by men and repeated physical contact principally by and/or abrasive dentiwomen. toothbrushes frices.1,7,10,18 For abfraction, it has been postulated that the cervical fulcrum area of a tooth is subject to unique stress, torque and moments resulting from occlusal function, bruxing and parafunctional activity.16,19-25 These flexural forces can act to disrupt the normal ordered crystalline structure of the thin enamel and underlying dentin by cyclic fatigue, leading to cracks, chips and ruptures.16,24-31 Stress corro-

JADA, Vol. 133, June 2002 Copyright 2002 American Dental Association. All rights reserved.

725

R E S E A R C H

sion and piezoelectric effects also have been theoas much as possible, of the investigators techrized to have an effect.14 niques, criteria and procedures. Evaluation caliNo single mechanism is adequate to explain all bration also included using standardized models occurrences of NCCLs. Their etiology likely is and photographs depicting the range of possible multifactorial in nature; a combination of all observations. Patients were screened initially to these factors is responsible to varying meet the study entry criteria (below) and, if qualidegrees.9,10,15,21,31-36 It has not been clearly identified, were enrolled in the study for the evaluation fied as to whether any one process is more responvisit. Qualified patients were recruited in the sible for lesion initiation or for progression, or order in which they presented themselves for the vice versa. Initiation of breakdown by one process screening session, thus forming a convenience can make the tooth more susceptible to damage sample. by the other processes, perhaps in a synergistic Procedure. The evaluative tools used were a manner.3,14,15,21,33 One factor may predominate over mouth mirror, explorer and periodontal probe. Air another in a given patient, leading to the varied from the air-water syringe was used to adminmorphological presentations.3,31 ister the thermal sensitivity test. The patients There are data to indicate that were preselected for severity of their occlusion, saliva, age, sex, diet and condition, being defined as having Knowing the parafunctional habits are factors at least three NCCLs that required characteristics of that may be associated with restoration. The lesions were established lesions NCCLs.6,15,16,31,37 Knowing the chardeemed to require restoration if acteristics of established lesions they were more than 1 millimeter in gives clues and gives clues and guidance as to if depth, had patient-reported thermal guidance as to if and and when intervention is indicated, sensitivity or both. The investigawhen intervention is and likely progression of lesions if tors screened all teeth in each subindicated, and likely left untreated. Identifying affected ject, selecting the largest and/or progression of lesions most sensitive lesions if there were teeth and susceptible patients if left untreated. enables judicious case selection for more than three teeth that qualified. They chose no more than three treatment and allows modification teeth per patient to minimize of treatment protocols, if patient-related effects that might distort the necessary.3,15,16 The ability to better estimate sucstudy results and yet obtain a sizeable number of cess and longevity of restorations will be a factor lesions. The other inclusionary criteria were that in treatment decision making, or at least provide subjects be older than 18 years of age and in good more realistic expectations of outcomes. This general health (American Society of Anesthesiarticle describes a study in which we examined ology Classification I or II).38 Patients were the characteristics of NCCLs, the teeth and the patients affected to determine trends, risk factors excluded if they exhibited active, untreated periand indicators for intervention and prognosis. odontal disease; had rampant, uncontrolled caries; experienced xerostomia; were undergoing METHOD AND MATERIALS orthodontic treatment or bleaching treatment; or Design. This study was a clinical survey of the were using supplemental fluoride. baseline pretreatment data, with descriptive We selected a total of 57 subjects, resulting in analysis and correlational analysis of lesion char171 teeth with NCCLs that were evaluated. acteristics, tooth location and patient demoLesions were characterized by shape, dimensions, graphics. While acknowledging that NCCLs are sensitivity, extent of sclerosis and occlusion.2,15,31,39-41 multifactorial in nature, this investigation We also noted tooth location and recorded patient approached the NCCL as a single entity to deterdemographic information. These factors have been mine if there are any trends, risk factors or etioidentified as possible covariables in NCCL formalogic covariablesin effect, to suggest a retrospection, and in adhesion of Class V resin-based tive analysis of the outcome. Patient screening composite restorations.42 The evaluation was perand evaluation of all teeth with NCCLs, identified formed after the investigator removed any debris, visually or tactilely, were performed by two clinplaque or surface layer that obscured inspection, ical investigators (T.C.A. and X.L.). We performed by means of gentle toothbrush agitation and use of an initial calibration to ensure standardization, floss. The investigator took quadrant impressions
726 JADA, Vol. 133, June 2002 Copyright 2002 American Dental Association. All rights reserved.

R E S E A R C H

of the tooth of interest using vinylpolysiloxane and poured stone models to have a three-dimensional record of evaluated teeth and respective NCCLs. Color photographs of each lesion also were taken for documentation. Evaluation criteria. Shape. The investigators evaluated the shape of the lesion by visually inspecting the vertical buccolingual cross section (Figure 1). Being that NCCLs are a saucer or a wedge shape, they categorized the acute angle formed by the occlusal and gingival walls on an ordinal scale as < 45 degrees, 45 to 90 degrees, 90 to 135 degrees and > 135 degrees. Dimensions. The investigators measured lesions dimensions by using a periodontal probe. They categorized the axial depth, judged by the estimated ideal buccal contour compared to adjacent or contralateral normal teeth to the most axial portion of the lesion, on an ordinal scale as 1 to 2 mm, 2 to 3 mm, 3 to 4 mm and > 4 mm (Figure 2). The occlusogingival width is the vertical distance at the widest extent between the occlusal and gingival margins of the lesion, also categorized on an ordinal scale as 1 to 2 mm, 2 to 3 mm, 3 to 4 mm and > 4 mm (Figure 2). Extent of sclerosis. The investigators evaluated the extent of sclerosis by visual inspection and tactile feedback with a dental explorer, to determine extent of discoloration (yellow or brown), glassy appearance (shiny, hard or smooth) and translucency or transparency of the enamel/ dentin. They categorized their observations on an ordinal scale as none, mild, moderate or heavy (as described in Duke and colleagues39,43) (Figure 3). Sensitivity. The investigators evaluated sensitivity by applying a blast of air from an air-water syringe at a distance of approximately 1 inch away, with adjacent teeth under rubber dam isolation (Figure 4). The application of air was for a maximum of five seconds. The subjective patient response of level of sensitivity (discomfort) was recorded on a 10-point continuous visual-analog ratio scale, ranging from no discomfort (a score of zero) to extreme discomfort (a score of 10). Occlusion. Occlusion was observed by visual inspection. Observations were in multiple categories of nominal scales of Angles classification (Class I, II and III), excursive guidance (canine guidance, group function or a mixture), wear facets (absent or present) and ordinal scale of mobility (Lindhe44 Type I, II and III). Angles classification and guidance were examined in the

Figure 1. Angular shape of cervical lesion in buccolingual cross section. A. Wedge or V shape. B. Saucer or C shape.

Figure 2. Dimensions of cervical lesion in buccolingual cross section. A. Horizontal depth. B. Vertical width.

Figure 3. Cervical lesions in a quadrant demonstrating the range of sclerosis: none (no. 22), mild (no. 21), moderate (no. 20) and heavy (no. 19).

Figure 4. Air sensitivity testing of a single tooth isolated with a rubber dam.

JADA, Vol. 133, June 2002 Copyright 2002 American Dental Association. All rights reserved.

727

R E S E A R C H

TABLE 1

LESION CHARACTERISTICS IN TEETH STUDIED.

CHARACTERISTIC Shape (Degree of Angle) < 45 45-90 90-135 > 135 Axial Depth (mm*) 1-2 2-3 3-4 >4 Occlusogingival Width (mm) 1-2 2-3 3-4 >4 Sclerosis None Mild Moderate Heavy Sensitivity None/Mild (score 0-3) Moderate (score 4-6) Extreme (score 7-10) Angles Classification I II III IV Guidance Canine Group Mixed Wear Facets Absent Present Mobility 0 I II III * mm: Millimeters. NO. (%) OF TEETH AFFECTED

11 (6) 64 (37) 64 (37) 32 (19) 156 (91) 13 (8) 2 (1) 0 84 (49) 43 (25) 29 (17) 15 (9) 37 93 38 3 (22) (54) (22) (2)

Patient demographic information. Patients were grouped by age on an ordinal scale: 21 to 40 years, 41 to 60 years and 61 to 80 years. Patients sex was recorded on a nominal scale as either male or female. This information was selfreported by the subject and not independently verified.
DATA COLLECTION AND ANALYSIS

125 (73) 24 (14) 22 (13) 129 (75) 25 (15) 17 (10) 0 68 (40) 63 (37) 40 (23) 30 (18) 141 (82) 70 (41) 100 (58) 1 (1) 0

For each evaluation criterion, the statistical expert on the research team (L.M.) calculated the category frequencies. He assessed pairwise associations between the evaluation criteria by logistic regression analysis using generalized estimating equations to account for the possible dependence between multiple lesions within a patient. All tests were performed at a .05 significance level.
RESULTS

quadrant of the tooth in question. Wear facets were identified visually by flat, smooth surfaces on the cusp ridges, triangular ridges or inclined planes of the involved tooth. The investigators measured tooth mobility using pressure exerted between the ends of two mirror handles. Tooth location. The investigators noted the position of the teeth, anterior or posterior, maxillary or mandibular, on a nominal scale. They also categorized the teeth as first molars, second premolars, first premolars, canines, lateral incisors or central incisors.
728

Table 1 shows the breakdown of the lesion characteristics of 171 teeth in 57 subjects (three lesions per subject); Table 2 shows the tooth location information; and Table 3 shows the subject demographic information. The angular shape of the lesions tended to be roughly right-angled (90 degrees), with 74 percent in the 45- to 135-degree range and only 6 percent less than 45 degrees. The axial depth tended to be shallow, with 91 percent being in the 1 to 2 mm range and only 1 percent greater than 3 mm. The occlusogingival width tended to be narrow, with 49 percent in the 1- to 2-mm range and 42 percent in the middle 2- to 4-mm range. Extent of sclerosis tended to be rated mild (54 percent); an equivalent number was

JADA, Vol. 133, June 2002 Copyright 2002 American Dental Association. All rights reserved.

R E S E A R C H

TABLE 2 split between none and moderate, and 3 percent LOCATION OF TEETH WITH NONCARIOUS CERVICAL of the sclerosis was LESIONS. judged to be heavy. For air sensitivity, 73 percent NO. (%) OF TEETH AFFECTED LOCATION felt either no or mild senAnterior vs. Posterior Location sitivity (in the 0-3 range), 52 (30) Anterior with roughly equal num119 (70) Posterior bers in the moderate (4-6 Maxillary vs. Mandibular Location range) and extreme (7-10 112 (65) Maxillary 59 (35) Mandibular range) categories. Occlusion was another Anterior Teeth 34 (20) Canines major characteristic that 5 (4) Lateral Incisors we examined extensively. 13 (7) Central Incisors Angles classification Posterior Teeth tended to be Class I, with 42 (25) First Molars 33 (20) Second Premolars a 75 percent majority; the 44 (26) First Premolars remaining proportions were almost equally TABLE 3 divided between Class II and Class III. Excursive SUBJECT DEMOGRAPHIC INFORMATION. guidance was 40 percent canine guidance, 37 perNO. (%) OF SUBJECTS CHARACTERISTIC cent group function, and Age (Years)* 23 percent a mixture of 9 (16) 21-40 the two. Wear facets were 34 (60) 41-60 13 (23) 61-80 present in most (82 percent) of the patients. Sex 31 (54) Male Tooth mobility tended to 26 (46) Female be not or only slightly mobile, with 99 percent * One subject opted not to reveal his or her age, leading to a total of 56 rather than 57. Type 0 or I, only 1 percent Type II and none Type III. Seventy percent of the NCCLs were on postemost lesions were not very deep, were not vertirior teeth and 30 percent on anterior teeth, cally wide and were approximately right-angled and 65 percent were on maxillary teeth and 35 in shape. It has been found that lesions are more percent on mandibular teeth. First premolars prevalent in the older population, and older (26 percent) and first molars (25 percent) were patients are more likely to have lesions that are affected the most often, followed by second predeeper, larger or both.7 This is not surprising molars (20 percent) and canines (20 percent). because older patients and their teeth have been NCCL incidence was least likely in central exposed to the pertinent etiologic factors for a incisors (7 percent) and lateral incisors (4 permuch longer period than younger patients (and cent). Patients tended to be older, with 60 pertheir teeth), and thus should be expected to have cent in the group aged 41 to 60 years and 23 permore lesions, and of greater severity.15,24 In addicent in the group aged 61 to 80 years. Only 16 tion, older populations are more likely to have percent were in the group aged 21 40 years. Subgingival recession and bone loss, with more root jects overall age range was 29 to 75 years, with surface and cementum exposure, increasing the a mean of 51 ( 13 standard deviation). Subjects risk of cervical lesions.45,46 In this study, 83 perwere almost equally divided between male (54 cent of the subjects were older than 40 years of percent) and female (46 percent). age, and 23 percent were older than 60 years, and yet axial depththe primary diagnostic criterion DISCUSSION for restoring NCCLswas greater than 2 mm in Shape and dimensions. It would appear that only 9 percent of these cases. This might suggest

JADA, Vol. 133, June 2002 Copyright 2002 American Dental Association. All rights reserved.

729

R E S E A R C H

that NCCL progression is a slow process, that cyclic lateral and compressive forces are exerted early restorative intervention may not be necesat the cervical fulcrum area of the teeth, as occurs sary, and that monitoring and re-evaluation may during chewing. More subjects were in group be acceptable, even appropriate, for small, asympfunction than in canine guidance, with more tomatic incipient NCCLs. opposing tooth contact in the former than the Sclerosis and sensitivity. The majority of latter. Excursive tooth contacts can exert powlesions were sclerosed to some extent, mostly in erful tensile and shearing forces, especially if the mild category. Contrary to what might be there are occlusal interferences. Thus the concept expected for such macroscopic loss of tooth strucof freedom in centric should diminish NCCL forture, sensitivity was mostly minimal or nonexismation, whereas bruxing, clenching and other tent. Since NCCL development tends to be a slow, parafunctional habits that increase the magnichronic process that occurs over an extended tude of cervical stress would increase NCCL forperiod, it was not surprising to find sclerosis and mation.31 Indirect evidence is provided by the lack of sensitivity. Secondary dentin, occlusion of finding that bruxers have a greater incidence of open dentinal tubules, pulpal NCCLs than nonbruxers.19,20 Overretreat and other natural tooth prowhelmingly, wear facets, a sign of tective measures have slowly stressful occlusion, were present Posterior teeth were adapted to the noxious stimuli, (rather than absent) on teeth with more likely to exhibit thereby minimizing symptoms and NCCLs, providing support for noncarious maintaining pulpal integrity.43 This occlusal forces and flexure as causal cervical lesions, does correspond to the expectation factors. The presence of wear facets possibly owing to the has been a common finding with that logically there should be an inverse relationship between sclecervical lesions.31,32 The theory of fact that greater rosis and sensitivity, with 76 perocclusal loading is given further occlusal forces and cent of lesions classified with mild support by the finding that almost more lateral forces or moderate sclerosis, and 73 perall (99 percent) of NCCL teeth had are exerted in the cent with mild or no sensitivity. little or no mobility. If teeth are posterior teeth. These findings are in accord with mobile, then concentration of forces other reports that NCCLs generally at the cervical area could not occur. exhibit a lack of thermal sensiOthers also have found a definite tivity.2,31 From a restorative standpoint, with the negative correlation between tooth mobility and presence of sclerosis and absence of sensitivity, presence of an NCCL.4 preparation design may require additional retenTooth location. Posterior teeth were more tion and alleviated sensitivity should be a routine likely to exhibit NCCLs, possibly owing to the outcome. However, one should bear in mind that fact that greater occlusal forces and more latsclerosis is a difficult category to measure, in that eral forces are exerted in the posterior teeth. it is fraught with subjectivity and susceptible to Maxillary teeth seem more prone to NCCLs, inter- and intraexaminer variation on features possibly owing to the lingual tilt. Premolars, in such as discoloration, smoothness and transluparticular first premolars, appeared in our cency. Sensitivity to cold air is also a difficult study to have the highest prevalence of NCCLs, measurement, relying solely on the subjects subwhereas incisors, especially laterals, have the jective response and perception of pain severity. lowest prevalence. Prevalence in first molars Occlusion. There is an increasing amount of also was high, second only to first premolars. evidence that occlusal factors, involving repeated These findings confirm those of many previous occlusal stresses and tooth flexure, play a signifistudies, that posterior maxillary teeth, specificant role in NCCL etiology. The majority of cally first premolars and first molars, are most affected teeth (75 percent) were in Class I occlususceptible to cervical lesions.1,2,10,11,15,31,36,47 This sion (considered the most desirable, normal has been attributed to the greater occlusal occlusion), which concurs with results of other forces in the posterior teeth or to natural relastudies.31 With a Class I occlusion, maximal intertive anatomical morphology of the teeth, periarch tooth contact in centric occlusal function can odontium and vestibule.3,10,31,32 Other possibilioccur during maximum intercuspation. The more ties are the natural progressive development of contact there is between opposing teeth, the more group function from anterior to posterior or the
730 JADA, Vol. 133, June 2002 Copyright 2002 American Dental Association. All rights reserved.

R E S E A R C H

relative accessibility of toothbrushes and 43 subjects with 178 lesions demonstrated strikbrushing mechanics.3,10,31,32 ingly analogous results to ours, with almost idenSubject demographics. As with other tical frequency distributions. However, it must be studies,7,10,15,18,36,47 the prevalence of NCCLs noted that the measurement criteria in our study increased with increasing age; the majority of this were post hoc data collected after the fact and study population was older than 40 years of age, provide only indirect evidence of NCCL formawith a mean age of 51 years. As mentioned prevition. Our study was conducted on a very specific ously, older patients who have been exposed to the subset of patients with NCCLs, namely those etiologic factors for a longer time, coupled with with at least three lesions of defined severity. The marginal tissue or gingival recession, are at data may not reflect a true random survey of the greater risk for cervical lesions and, thus, more entire general population of people with NCCLs. likely to have more lesions and lesions of greater There also may be unknown possible sampling severity.15,24,44 In addition, older subjects are more bias of self-selected subjects of which we are likely to have fewer teeth to bear the occlusal load, unaware, since they essentially volunteered to be with a loss of the protective mechanisms of natural recruited into the study population. Nonetheless, dentition and diminished quantity since a sizable number of patients and quality of saliva.13,15,16,37,48 Also, were recruited, and they had a compositional and microstructural high incidence of lesions, this can Lesions of smaller changes to enamel and dentin associbe considered a sufficiently ranwidths were ated with the aging process may domized group, and these lesions associated with render the tooth structure more susaccurately represent the true greater sensitivity, ceptible to lesion formation; the prenature of the NCCL. cise role of these changes still is not There are many misconceptions perhaps owing to fully understood.16,26,39,45 Reflecting of the causes of such lesions and less sclerosis. past findings, our findings do not substantial differences among denappear to point to any sex difference tists in the recognition and treatin NCCL prevalence.7,10 ment of cervical lesions.5 The Pairwise associations. Geometry dictates results from this study should aid the clinician in that lesion shape (angle) should be associated identifying and deciding which teeth require with occlusogingival widthand indeed, the restoration and which patients are the best candismaller the shape, the smaller the vertical dimendates for the procedure. It also helps in predicting sion. Perhaps not surprisingly, occlusogingival the likely progression of lesions if they are left width also was associated with axial depth; the untreated. The treatment plan may include presmaller the vertical width, the smaller the axial treatment adjustment or alteration of occlusion. depth. Lesions with smaller depth were associIn selecting restorative materials, the clinician ated with less sclerosis, reflecting more sclerosis should look for a low modulus of elasticity, resisat the later stage of a lesion. Lesions of smaller tance to wear and ability to withstand acid dissowidths were associated with greater sensitivity, lution.3,16 Preparation design may need to be perhaps owing to less sclerosis. As expected, changed, depending on whether adhesive retenlesions with no or mild sclerosis were more sensition is adequate or if mechanical retention is nective than those with moderate or heavy sclerosis. essary.3,16 Preventive interventions include use of Also not surprising was our finding that wear a protective nightguard, changing toothbrushes facets were more likely with group function and and dentifrices, and use of neutralizing mixed guidance than with canine guidance. Posmouthrinses. A patients behavior and habits may terior teeth were more likely to have less require modifications such as changes in diet; mobility, probably owing to their multirooted changes in brushing technique, force and frenature. Mandibular lesions were more likely to be quency; and reduction of clenching or bruxing.5,48 on posterior teeth than were maxillary lesions, The same characteristics that have been identiconsistent with the rarity of finding mandibular fied as associated with NCCL formation have anterior NCCLs. Associations do not appear to be been hypothesized to be predictive of restoration transitive; in other words, two related associasuccess or failure.19 A prospective, longitudinal, tions do not necessarily imply a third association. randomized clinical trial of Class V cervical A similar study by Mayhew and colleagues31 of restorations on these lesions can yield insights as
JADA, Vol. 133, June 2002 Copyright 2002 American Dental Association. All rights reserved. 731

R E S E A R C H

to what determines clinical success rates and the variables that are important to this outcome.
CONCLUSION

We examined 57 subjects who had a total of 171 teeth with multiple Dr. Aw is an assistant professor, Division of NCCLs of significant size. We Operative Dentistry, found that NCCL progression was Department of Restorative Dentistry, Univera slow process, with most lesions sity of Washington, School of Dentistry, remaining quite small even in D-770 Health Sciences subjects of advanced age. Most Building, Box 357456, NCCLs were sclerotic, leading to Seattle, Wash. 981957456, e-mail tcaw@ diminished sensitivity. Occlusion u.washington.edu. tended to be Class I, with a group Address reprint requests to Dr. Aw. function excursive guidance, with a preponderance of wear facets and little or no mobility, lending indirect evidence to an occlusal stress/tooth flexure etiology. Cervical lesions were more common with posterior maxillary teeth. Premolars as a group, and first premolars in particular, were most likely to have cervical lesions, and incisors, especially lateral incisors, were the least likely. NCCLs were more likely in older patients, but prevalence by sex was equivalent. These characteristics and covariables help identify whether to intervene and how to treat NCCLs, as well as which teeth and patients are more susceptible. I
The authors acknowledge the support of 3M ESPE Corp., St. Paul, Minn., and Coltne AG, Altsttten, Switzerland, for providing materials and funding for this study. The authors acknowledge the support of the study site, the University of Washington Regional Clinical Dental Research Center, for the facilities, equipment and particularly the efforts of the staff members, all of which have been instrumental in the conduct of this clinical trial. 1. Hong FL, Nu ZY, Xie, XM. Clinical classification and therapeutic design of dental cervical abrasion. Gerodontics 1988;4(2):101-3. 2. Hollinger JO, Moore EM Jr. Hard tissue loss at the cementoenamel junction. J N J Dent Assoc 1979;50(4):27-31. 3. Osborne-Smith KL, Burke FJ, Wilson NH. The aetiology of the non-carious cervical lesion. Int Dent J 1999;49(3):139-43. 4. Hand JS, Hunt RJ, Reinhardt JW. The prevalence and treatment implications of cervical abrasion in the elderly. Gerodontics 1986;2(5):167-70. 5. Bader JD, Levitch LC, Shugars DA, Heymann HO, McClure F. How dentists classified and treated non-carious cervical lesions. JADA 1993;124(5):46-54. 6. Eccles JD, Jenkins WG. Dental erosion and diet. J Dent 1974;2(4):153-9. 7. Bergstrom J, Lavstedt S. An epidemiologic approach to toothbrushing and dental abrasion. Community Dent Oral Epidemiol 1979;7(1):57-64. 8. Balanko M, Jordan RE. Gingivally submerged cervical erosion lesion: a clinical problem. J Esthet Dent 1990;2(4):104-8. 9. Poynter ME, Wright PS. Tooth wear and some factors influencing its severity. Restorative Dent 1990:6(4):8-11. 10. Radentz WH, Barnes GP, Cutright DE. A survey of factors possibly associated with cervical abrasion of tooth surfaces. J Periodontol 1976;47(3):148-54. 11. Xhonga FA, Valdmanis S. Geographic comparisons of the incidence of dental erosion: a two centre study. J Oral Rehabil 1983;10:269-77.

Dr. Lepe is an associate professor, Division of Fixed Prosthodontics, Department of Restorative Dentistry, University of Washington, School of Dentistry, Seattle.

Dr. Johnson is a professor, Division of Biomaterials and Research, Department of Restorative Dentistry, University of Washington, School of Dentistry, Seattle.

Dr. Mancl is a research assistant professor, Department of Dental Public Health Sciences, University of Washington, School of Dentistry, Seattle.

12. Schaffner M, Suter P, Hotz P, Lussi A. Epidemiology of dental erosion and toothbrush abrasion (abstract 2205). J Dent Res 1988;67:388. 13. Jarvinen VK, Rytomaa II, Heinonen OP. Risk factors in dental erosion. J Dent Res 1991;70:942-7. 14. Grippo JO, Simring M. Dental erosion revisited. JADA 1995;126:619-30. 15. Levitch LC, Bader JD, Shugars DA, Heymann HO. Non-carious cervical lesions. J Dent 1994;22(4):195-207. 16. Lee WC, Eakle WS. Stress-induced cervical lesions: review of advances in the past 10 years. J Prosthet Dent 1996;75(5):487-94. 17. Pindborg JJ. Pathology of the dental hard tissues. Philadelphia: Saunders; 1970. 18. Bergstrom J, Eliasson S. Cervical abrasion in relation to toothbrushing and periodontal health. Scand J Dent Res 1988;96:405-11. 19. Heymann HO, Sturdevant JR, Bayne S, Wilder AD, Sluder TB, Brunson WD. Examining tooth flexure effects on cervical restorations: a two-year clinical study. JADA 1991;122(6):41-7. 20. Xhonga FA. Bruxism and its effect on the teeth. J Oral Rehabil 1977;4(1):65-76. 21. Sognnaes RF, Wolcott RB, Xhonga FA. Dental erosion I: erosionlike patterns occurring in association with other dental conditions. JADA 1972;84:571-6. 22. Xhonga FA, Wolcott RB, Sognnaes RF. Dental erosion II: clinical measurements of dental erosion progress. JADA 1972;84:577-82. 23. Thresher RW, Saito GE. The stress analysis of human teeth. J Biomech 1973;6:443-9. 24. Lee WC, Eakle WS. Possible role of tensile stress in the etiology of cervical erosive lesions of teeth. J Prosthet Dent 1984;52:374-80. 25. Yettram AL, Wright KW, Pickard HM. Finite element stress analysis of the crowns of normal and restored teeth. J Dent Res 1976;55:1004-11. 26. Goel VK, Khera SC, Ralston JL, Chang KH. Stresses at the dentinoenamel junction of human teeth: a finite element investigation. J Prosthet Dent 1991;66:451-9. 27. Rasmussen ST, Patchin RE, Scott DB, Heuer AH. Fracture properties of human enamel and dentin. J Dent Res 1976;55(1):154-64. 28. Braem M, Lambrechts P, Vanherle G. Stress-induced cervical lesions. J Prosthet Dent 1992;67:718-22. 29. McCoy G. The etiology of gingival erosion. J Oral Implantol 1982;10:361-2. 30. Powers JM, Craig RG, Ludema KC. Frictional behavior and surface failure of human enamel. J Dent Res 1973;52:1327-31. 31. Mayhew RB, Jessee SA, Martin RE. Association of occlusal, periodontal, and dietary factors with the presence of non-carious cervical dental lesions. Am J Dent 1998;11(1):29-32. 32. Bader JD, McClure F, Scurria MS, Shugars DA, Heymann HO. Case-control study of non-carious cervical lesions. Community Dent Oral Epidemiol 1996;24(4):286-91. 33. Konig K. Root lesions. Int Dent J 1990;40:283-8. 34. Eccles JD. Dental erosion of nonindustrial origin: a clinical survey and classification. J Prosthet Dent 1979;42:649-53. 35. Spranger H. Investigation into the genesis of angular lesions at the cervical region of teeth. Quintessence Int 1995;26(2):149-54. 36. Piotrowski BT, Gillette WB, Hancock EB. Examining the prevalence and characteristics of abfractionlike cervical lesions in a population of U.S. veterans. JADA 2001;132:1694-701. 37. Davis WB, Winter PJ. Dietary erosion of adult dentine and enamel: protection with a fluoride toothpaste. Br Dent J 1977;143(4):116-9. 38. American Society of Anesthesiologists. The ASA physical status

732

JADA, Vol. 133, June 2002 Copyright 2002 American Dental Association. All rights reserved.

R E S E A R C H

classification system. Available at: www.asahq.org/ProfInfo/ PhysicalStatus.html. Accessed Sept. 30, 2001. 39. Duke ES, Lindemuth J. Variability of clinical dentin substrates. Am J Dent 1991;4:241-6. 40. Harnirattisai C, Inokoshi S, Shimada Y, Hosoda H. Adhesive interface between resin and etched dentin of cervical erosion/abrasion lesions. Oper Dent 1993;18(4):138-43. 41. Shimada Y, Harnirattisai C, Inokoshi S, Burrow MF, Takatsu T. In vivo adhesive interface between resin and dentin. Oper Dent 1995;20(5):204-10. 42. Bayne SC, Heymann HO, Sturdevant JR, Wilder AD, Sluder TB. Contributing co-variables in clinical trials. Am J Dent 1991;4(5):247-50. 43. Duke ES, Lindemuth J. Polymeric adhesion to dentin: contrasting substrates. Am J Dent 1990;3:264-70.

44. Lindhe J. Examination of patients with periodontal disease. In: Lindhe J. Textbook of clinical periodontology. 2nd ed. Copenhagen, Denmark: Munksgaard; 1989. 45. Ketterl W. Age-induced changes in the teeth and their attachment apparatus. Int Dent J 1983;33:262-71. 46. Alvares OF, Johnson BD. The aging of the periodontium. In: Wilson TG, Kornman KS, eds. Fundamentals of periodontics. Chicago: Quintessence Publishing; 1996:175. 47. Lussi A, Schaffner M, Hotz P, Suter P. Dental erosion in a population of Swiss adults. Community Dent Oral Epidemiol 1991;19: 286-90. 48. Woltgens JH, Vingerling P, de Blieck-Hogervorst JM, Bervoets DJ. Enamel erosion and saliva. Clin Prev Dent 1985;7(3):8-10.

JADA, Vol. 133, June 2002 Copyright 2002 American Dental Association. All rights reserved.

733