Background: Various cross-sectional forms of non-carious cervical lesions have been described but no formal morphological classification system has been developed. The aims of this study were to describe the spectrum of common morphological forms of non-carious cervical lesions observed within a large sample of extracted human permanent anterior teeth and to develop a system for classification based on morphological features.
Methods: Over 15 000 extracted permanent anterior teeth were examined macroscopically under illumination at 2× magnification. Well-defined, descriptive categories were formed, based on observable non-carious cervical lesion features and using terminology currently reported in the literature. The lesions were then sorted into these categories.
Results: Five hundred and forty-two non-carious cervical lesions were identified on 15 289 teeth extracted in the early to mid 20th century, representing a frequency of 3.5%. The main categories developed were “shallow”, “concave”, “wedge-shaped”, “notched”, and “irregular”.
Conclusions: The new morphological classification system provides an alternative to presently used systems based on aetiology, and should facilitate future research on non-carious cervical lesions.
Non-carious cervical lesions (NCCLs) involve loss of hard tissue at the cervical third of the dental crown and subjacent root surface, through processes unrelated to caries.1 The prevalence of NCCLs reported in epidemiological studies varies from 0% to 85%, with the wide range of estimates being attributed to different study design parameters, such as the methodology used for determining the presence of NCCLs, samples sizes and the populations studied.2,3 The aetiology of NCCLs is commonly multifactorial, with combinations of distinct processes, including abrasion, corrosion (erosion) and possibly abfraction most often operating to varying degrees.4–6 There is continued debate relating to the multifactorial aetiology of NCCLs, predominantly in relation to which processes are operating.2,4,5
NCCLs may cause problems such as tooth sensitivity, increased plaque retention, poor aesthetics and compromised pulp vitality, and they are most commonly diagnosed by detailed history taking, visual examination and tactile examination. They are managed by addressing any identifiable underlying causes, the application of desensitizing agents, and the placement of restorations, but prevention and management can be difficult if their aetiology is unclear.7 In many instances, NCCLs are just monitored.2,8–10
Previous studies have investigated different features of NCCLs and the teeth which exhibit them through various means, including measurement, subjective observation and scanning electron microscopy.7,8,10–23 The appearance of NCCLs, based on their cross-sectional shape, has been described but no formal morphological classification system has been developed. Lack of agreement regarding the most appropriate term(s) to use when describing a given morphological form of NCCL has added to the confusion in this controversial area of dentistry.11
The aims of this study were to describe the spectrum of common morphological forms of NCCLs observed within a large sample of extracted human permanent anterior teeth and to develop a simple and logical descriptive system for their classification based on morphological features.
Materials and methods
The first step of this study involved examining macroscopically a sample of 15 289 extracted permanent anterior teeth stored in the Murray Barrett Laboratory in the Adelaide Dental Hospital. To the best of the authors’ knowledge, this sample was composed of teeth extracted in the early to mid-20th century in the Adelaide Dental Hospital clinics.
All cervical surfaces of individual anterior teeth, held by their incisal and apical extremities, were carefully examined under illumination at approximately 2× magnification. The first step involved examining approximately 2000 teeth. These teeth were sorted into two categories, based on whether or not they had an NCCL. A limited number of teeth exhibited cervical damage on their facial and lingual surfaces associated with the process of extraction. These lesions were not studied further. As the NCCLs were examined, cross-sectional features were noted and morphological categories were developed. Following this, all incisors and canines were assessed and sorted into categories.
The intra-examiner reliability of categorizing the morphological forms of NCCLs was tested using double determinations. The reliability tests involved the random selection of teeth from each of the NCCL categories, providing a total of 32 teeth for the test. For testing intra-examiner reliability, scoring of teeth was performed one day apart. The examiner was blind to previous results and the exact number of teeth from each category involved in the test.
In total, 542 NCCLs were observed giving a prevalence of 3.5%. NCCLs were found most commonly on facial surfaces but some were evident lingually. They were found more commonly on incisors compared with canines. Based on the cross-sectional shape of the NCCLs observed, five main categories were created (Fig 1):
(1) Shallow NCCLs – lesions that had a shallow depth (less than or equal to 0.5 mm) but their breadth and height dimensions were greater than 1.0 mm. These lesions exhibited minimal dentine exposure. The cross-sections of shallow NCCLs were either flat or curved (Fig 2).
(2) Concave NCCLs – lesions that displayed a concave cross-section and had a depth greater than 0.5 mm. Concave NCCLs did not demonstrate a well-defined internal line angle. Instead, the floor of these lesions was continuously curved (Figs 2 and 3).
(3) Wedge-shaped NCCLs – lesions that had a wedge-shaped cross-section. Wedge-shaped lesions demonstrated a well-defined internal line angle and flat internal floors (Figs 2 and 4). When multiple wedge shapes were evident within the boundaries of the NCCL, the lesion was considered to be an irregular NCCL (see below).
(4) Notched NCCLs – lesions that had a very short corono-apical height (0.5–1.0 mm) relative to a longer mesio-distal length (4.0–6.0 mm). Their cross-sections were either wedge-shaped or concave. Wedge-shaped and concave NCCLs did not demonstrate these unique dimensional proportions (Fig 2).
(5) Irregular NCCLs – lesions that could not be categorized into one of the above NCCL categories, as they had an irregular cross-section. These lesions exhibited either angular features (e.g., multiple wedge shapes), curved features (e.g., wave-like in cross-section) or a combination of angular and curved features (Fig 2).
The percentage concordance in scoring 32 teeth with NCCLs on two separate occasions was 96%, indicating a high level of intra-observer reliability.
The spectrum of common morphological forms of NCCLs observed within a large sample of extracted human permanent anterior teeth was described and a morphologically-based, descriptive classification system for NCCLs developed. Where appropriate, the classification system has incorporated terms used throughout the literature to describe the appearance of NCCLs.3,7,16,21 We believe it is important to be consistent in use of terminology to describe NCCLs as multiple synonymous terms have tended to lead to confusion and misunderstanding in this area of dentistry (Table 1).
Table 1. Examples of categories in the new morphological NCCL classification system with synonymous terms provided in the literature3,7,16,19,20,22
Morphological NCCL category
Shallow grooves, grooved, gingival notching, deep notches
Irregular NCCLs identified in this study have not been described previously. Previous researchers may have concluded that irregular NCCLs were not present in large enough numbers to describe them as a unique morphological form and they may have classified them according to their dominant morphology (e.g., wedged). Although there may be many forms of irregular NCCLs, it was considered important to recognize them as a distinct group as they may have complicated aetiologies and thus require special management.
NCCLs are commonly classified according to their assumed aetiology.8 However, the predominant component of this aetiology is not always obvious. Therefore, classification of NCCLs based on aetiology can be difficult and hence the value of this new descriptive, morphological system. Even when the main aetiological factor can be determined, a morphological classification adds additional information of potential value to the clinician or researcher.
Although there is no clear evidence about how the different forms of NCCLs described in this new system may relate to one another, one can postulate that they reflect different stages of progression of the lesion. For example, shallow NCCLs may develop concave features. Furthermore, notched NCCLs may progress to become wedge-shaped. Given the lack of firm evidence to support these suggestions, longitudinal clinical studies designed to assess how the various forms of NCCLs change in morphology over time would be particularly valuable. These studies could also clarify whether particular morphological forms of NCCLs are associated with particular contributions of aetiological factors. If certain NCCL morphologies are consistently shown to be associated with particular aetiological factors, then clinical diagnosis based on NCCL morphology would be greatly assisted. Future clinical studies may also clarify whether different NCCL morphologies are associated with different success rates when restored with a given material. If this type of evidence is found, then classifying the morphology of an NCCL could provide a guide to the most appropriate method of restoration.
Although intra-operator reliability in classifying the different forms of NCCLs was high, it was not perfect. This suggests that some inter-operator disagreement could also be anticipated. Like most classification systems, this system has an element of subjectivity and different oral health professionals may have different interpretations of the morphological NCCL categories. Nevertheless, the system is relatively simple to apply and the categories clearly defined, suggesting that it should be applicable to clinical situations.
An important future direction for applying this classification system would be to determine if it can be used for posterior teeth. Since it is difficult to access a large sample of modern extracted permanent posterior teeth, a cross-sectional clinical study would be valuable. This would also enable an assessment of the applicability of the system in clinical practice, including its intra- and inter-operator reliability.
The five main forms of NCCL that may be encountered in permanent anterior teeth can be classified as “shallow”, “concave”, “wedge-shaped”, “notched”, and “irregular”. The shallow category includes both flat and curved forms, while the irregular category includes lesions with angular and/or curved forms. This new classification system is based on morphological features and it provides an alternative to the more frequently used descriptions that are based on aetiology. Clinical research is now needed to determine how the various forms of NCCLs change over time and whether classifying them according to both their morphological appearance and their aetiology can facilitate diagnosis, prevention and treatment.
We thank Emeritus Professor Tasman Brown for his assistance in determining the origin of the sample of extracted teeth used in this research. We also thank Dr Sabrina Woodroffe for her assistance in coordinating the intra-operator reliability tests. This research was generously supported by an Australian Dental Research Foundation Undergraduate Research Grant.