Journal of Oral Rehabilitation 2008 35; 128–134
The prevalence and severity of non-carious cervical lesions in
a group of patients attending a university hospital in Trinidad
W. A. J. SMITH, S. MARCHAN & R. N. RAFEEK
The University of the West Indies, Trinidad, West Indies
SUMMARY Non-carious cervical lesions (NCCLs) are
often encountered in clinical practice and their
aetiology attributed to toothbrush abrasion, erosion
and tooth flexure. This paper aims to determine the
prevalence and severity of NCCLs in a sample of
patients attending a university clinic in Trinidad and
to investigate the relationship with medical and
dental histories, oral hygiene practices, dietary habits
and occlusion. Data were collected via a question-
naire and clinical examination. Odds ratios were
used to determine the association of the presence of
lesions and the factors examined. One hundred and
fifty-six patients with a mean age of 40Æ6 years were
examined of whom 62Æ2% had one or more NCCLs.
Forty five per cent of the lesions were sensitive to
compressed air. Younger age groups had a signifi-
cantly lower correlation with the presence of NCCLs
Introduction
Non-carious cervical lesions (NCCLs) refer to the loss of
tooth structure at the cervical aspects of teeth. The
management of these lesions can pose challenges to
restorative dentists in terms of providing a lasting
restoration, preventing the progression of existing
lesions and development of new lesions, managing
dentine sensitivity and restoring the aesthetics and
function of teeth with lesions. Many studies have
reported the prevalence of these lesions and a range of
5–85% was reported in one review (1). Other studies
have looked at the prevalence of teeth having lesions
(2, 3). The broad range suggests variations in the
definition of NCCLs or difficulties in detecting them.
It is generally agreed that the lesions are formed
due to a combination of abrasion, erosion and tooth
flexure (1). Toothbrush abrasion is one of the factors
ª 2007 The Author. Journal compilation ª 2007 Blackwell Publishing Ltd
School of Dentistry, Faculty of Medical Sciences,
than older age groups. Other significant factors
included patients who reported heartburn, gastric
reflux, headaches, bruxism, sensitive teeth and
swimming or had a history of broken restorations in
the last year. There was also significant correlation of
NCCLs in patients who brushed more than once a day
or used a medium or hard toothbrush. Patients with
vegetarian diets and those who reported consuming
citrus fruits, soft drinks, alcohol, yoghurt and vita-
min C drinks were associated with the presence of
lesions. Significant associations were also found in
patients with group function, faceting, clicking joints
or those who wore occlusal splints.
KEYWORDS: non-carious cervical lesions, tooth wear,
Trinidad, University of the West Indies
Accepted for publication: 1 April 2007
historically associated with the development of
NCCLs. Researchers have investigated the effects
of tooth brushing technique and force, frequency of
brushing, abrasive potential of the dentifrice and
bristle stiffness on the development of NCCLs but
with conflicting results (1, 4). Erosion is a well-known
cause of tooth surface loss and has been associated
with non-carious lesions (5, 6). The acid can come
from the stomach, the diet or other extrinsic sources.
Tooth flexure has also been described as a cause of
NCCLs (7) and the term abfraction coined to describe
the process where lesions are formed by tooth
structure breaking away due to the stresses generated
as a result of cuspal loading. There has been much
work done to substantiate this theory through photo-
elastic stress studies, strain gauge studies and finite
element stress analysis (8). However, there is little
clinical evidence to support this theory (4).
doi: 10.1111/j.1365-2842.2007.01763.x
 NON-CARIOUS CERVICAL LESIONS IN TRINIDAD 129

Some authors have attempted to classify non-carious Methods

lesions according to a single aetiological factor based on
the morphology of the lesion. Generally, shallow saucer Questionnaire
shaped lesions are attributed to an erosive aetiology;
A 32-item questionnaire was administered to a con-
wedged shaped or grooved lesions are attributed to
venience sample of new patients 16 years and older
abrasion and V-shaped lesions attributed to tooth
attending the UWI, School of Dentistry adult dental
flexure (1) (Fig. 1). However, it has been clearly
clinic between September 2002 and March 2003.
demonstrated that it is difficult to classify lesions in
Questions elicited basic demographics (name, age,
this manner based on studies which suggest that NCCLs
race), medical and dental history, oral hygiene or
have a multifactorial aetiology (9–11). Also, attempts to
parafunctional habits and dietary practices. Distinctive
diagnose NCCLs attributable to one single aetiological
questions in the questionnaire on medical history were
factor provide variable results (12). Despite this, some
included: frequency of suffering from gastric reflux,
authors have investigated NCCLs attributed to one
heartburn and vomiting. The items on parafunctional
aetiological factor based on morphology (3, 13, 14). The
habits included awareness of grinding or clenching
usefulness of these investigations remains debatable.
habits, whether an occlusal splint had ever been
There has been no data on the prevalence of NCCLs in
prescribed or worn and whether teeth or restora-
the Caribbean region. The prevalence of generalized
tions had fractured. The items on diet examined
tooth wear has been reportedly common in Trinidad and
the frequency of consumption of drinks known to have
has been associated with certain dietary patterns (15).
some acidic content, such as, citrus fruits, fruit juices,
The aim of this study was to determine the prevalence
yogurt and carbonated beverages (soft drinks). The
and severity of all forms of NCCLs in a sample of patients
items on oral hygiene included type of tooth-
attending an undergraduate teaching clinic at a Univer-
brush ⁄ toothpaste used and frequency of brushing.
sity hospital in Trinidad and to investigate the relation-
ship with medical and dental histories, oral hygiene
practices, dietary habits and occlusion. Examination

Two examining teams each consisting of an examiner

and scribe met for three calibration sessions prior to the
survey. Calibration was confirmed by comparing dupli-
cate examinations of 10 patients. Visual examinations
were conducted with the aid of a high intensity dental
light and dental mirror with the patients in a supine
position. All the teeth between and including the first
molars were examined in both arches for the presence
or absence of lesions.
The NCCL was scored according to the dentine
exposure and presence or absence of symptoms, as
follows: 0-no lesion present, 1-dentine visible, asymp-
tomatic, 2- dentine visible, sensitive to cold, 3-tertiary
dentine ⁄ pulp visible, asymptomatic, 4- tertiary den-
tine ⁄ pulp visible, sensitive to cold. In addition the
presence or absence of wear facets on both premolars
and the first molars in all quadrants and the type of
guidance (canine or group function) was noted.
Data were entered and analysed using SPSS version
11.0.1* and odds ratios calculated to show correlation

Fig. 1. A non-carious cervical lesion present on the labial of an

upper lateral incisor extending into tertiary dentine. *SPSS Inc., Chicago, IL, USA.

ª 2007 The Authors. Journal compilation ª 2007 Blackwell Publishing Ltd

130 W . A . J . S M I T H et al.

between postulated causative factors and the presence of
NCCLs.
Results
One hundred and fifty-six patients were examined with
ages ranging from 16 to 73 years, with a mean age of
40Æ6 years. Of these subjects, 33% were male and 67%
were female. Of all the patients examined 62Æ2% had at
least one NCCL. The relative percentages of teeth and
the different types of lesions examined are shown in
Table 1 and Figs 1 and 2.
Of the teeth examined, 13Æ1% had NCCLs. Most of
the lesions extended only into dentine: only 1% of the
lesions examined involved tertiary dentine. Of all the
lesions, 55% were asymptomatic while the remaining
45% were sensitive to compressed air.
Odds ratios were calculated using various groupings
of the sample to determine the association between the
presence of NCCLs and other factors. Younger age
groups had a significantly lower association of having
NCCLs than older age groups as shown in Table 2.
There was no association between the patient’s gender
and the presence of NCCLs.
Medical history
There were significant associations found between
NCCLs and patients who gave a history of heartburn,
gastric reflux and headaches. This is shown in Table 3.

Table 1. Non-carious cervical

Frequency (%)
lesions frequency table – reported
No lesion present 2653 (86Æ9) by tooth
Lesions present:
Dentine visible, asymptomatic 221 (55)%
Dentine visible, symptomatic 175 (44)%
Tertiary dentine visible, asymptomatic 0
Tertiary dentine visible, symptomatic 3 (1)%
Total 399 (100)% 399 (13Æ1)
Total number of teeth examined 3052 (100)

(a) 14 Percentages of upper teeth examined with non-carious cervical lesions

12

10 Dentine visible, asymptomatic

Dentine visible, sensitive
8 to cold
Tertiary dentine/pulp visible,
asymptomatic
6
Tertiary dentine/pulp visible,
sensitive to cold
4

0
UR6 UR5 UR4 UR3 UR2 UR1 UL1 UL2 UL3 UL4 UL5 UL6

(b) Percentages of lower teeth examined with non-carious cervical lesions

16

14

12
Dentine visible, asymptomatic
10 Dentine visible, sensitive
to cold
8
Tertiary dentine/pulp visible,
6 asymptomatic
Tertiary dentine/pulp visible,
4 sensitive to cold

0 Fig. 2. Proportion of teeth examined

LR6 LR5 LR4 LR3 LR2 LR1 LL1 LL2 LL3 LL4 LL5 LL6 with non-carious cervical lesions.

ª 2007 The Authors. Journal compilation ª 2007 Blackwell Publishing Ltd

 NON-CARIOUS CERVICAL LESIONS IN TRINIDAD 131

Table 2. The strength of association of age with the presence of who brushed once a day. Similarly, patients who used
non-carious cervical lesions expressed using odds ratio with 95% either a hard or medium brush had a stronger associ-
confidence intervals
ation with NCCLs than those who used a soft tooth
brush. Most patients could not recall the type of
Age groups Odds ratio 95% confidence interval
toothpaste they used and so these data were not
16–24 years vs. 25+ years 4Æ6 1Æ9, 11Æ0 included in the analysis.
16–50 years vs. 51+ years 4Æ1 1Æ7, 10Æ0
16–64 years vs. 65+ years 2Æ5 0Æ3, 22Æ9
Recreational sports
Table 3. The strength of association between various medical, Patients who reported swimming daily or weekly had
dental and other factors and non-carious cervical lesions expressed significantly higher associations with the presence of
as odds ratio with 95% confidence intervals
NCCLs compared to those who reported swimming
monthly or irregularly. There was no association of the
Group examined vs. Odds 95% confidence
Variable reference group ratio interval prevalence of NCCLs among patients who lived near a
beach or those who participated in outdoor sports.
Gastric reflux Daily or weekly vs. 1Æ2 0Æ3, 5Æ2
monthly or never
Heartburn Daily or weekly vs. 1Æ6 0Æ3, 8Æ7 Diet
monthly or never
Headaches Weekly vs. 1Æ4 0Æ8, 2Æ8 There was a significant association of NCCLs and
irregularly patients who were vegetarian or who reported the
Bruxism Bruxism vs. 1Æ0 0Æ5, 2Æ1
presence of citrus fruits, soft drinks, alcohol, yoghurt,
no bruxism
Clicking joint Clicking joint vs. 1Æ6 0Æ7, 3Æ4
chewing gum and effervescent vitamin C in their diets.
no clicking joint This is shown in Table 4.
Occlusal splint Occlusal splint wearer 2Æ5 0Æ3, 23Æ0 There was a significant association of NCCLs among
wearer vs. non-wearer groups of patients who reported of having citrus fruits
Patients Sensitive teeth vs. 2Æ0 1Æ0, 4Æ0 or soft drinks more than once a day versus once a day
complaining of normal sensitivity
or less often; daily or more often versus weekly or less
sensitive teeth
Broken filling Once a year vs. never 1Æ1 0Æ5, 2Æ4 often and weekly or more often versus monthly or
Tooth brushing Twice a day or more 4Æ7 1Æ4, 15Æ9 rarely. Similarly, there was a significant association of
often vs. once a day NCCLs among groups of patients who drank sports
Toothbrush type Soft vs. hard or medium 1Æ1 0Æ5, 2Æ2 drinks, alcohol or effervescent vitamin C drinks daily or
Swimming Daily or weekly vs. once 1Æ4 0Æ5, 4Æ2
more often versus weekly or less often and weekly
a month or less often
or more often versus monthly or less often. The data for
citrus fruits are presented as an example in Table 5.

Dental history

There were significant associations between NCCLs and

Table 4. The strength of association between the presence of
patients who were aware of bruxism (either grinding or
certain foods and beverages in the patients’ diet and non-carious
clenching) their teeth, who had a splint or mouth
cervical lesions expressed as odds ratio with 95% confidence
guard made for them, who complained of having intervals
sensitive teeth, who had a clicking temperomandibular
joint, or who had an incident of a broken filling in the 95% confidence
last year. The details of these findings are presented in Variable Odds ratio interval
Table 3. Vegetarian 6Æ7 0Æ8, 53Æ8
Citrus fruits 1Æ1 0Æ2, 6Æ7
Soft drinks 1Æ5 0Æ6, 4Æ3
Oral hygiene practices Alcohol 1Æ1 0Æ6, 2Æ2
Yoghurt 1Æ6 0Æ8, 3Æ1
Patients who brushed twice a day or more often had a
Effervescent vitamin C 1Æ3 0Æ6, 3Æ0
stronger association (OR 4Æ7) with NCCLs than those

ª 2007 The Authors. Journal compilation ª 2007 Blackwell Publishing Ltd

132 W . A . J . S M I T H et al.

Table 5. The strength of association between the frequency of morphology because studies have shown that most of the
intake of citrus fruits and non-carious cervical lesions expressed as lesions are multifactorial (10) and there is difficulty in
odds ratio with 95% confidence intervals
attributing one primary aetiological factor to a lesion
(4, 9). In addition, class V restorations or teeth restored
Group examined vs. Odds 95% confidence
reference group ratio interval with crowns were not included in the data. In a review
conducted by Levitch et al. (1), it was found that the
More than once a day vs. 1Æ1 0Æ4, 2Æ8
reported prevalence rates for NCCLs range from 5% to
once a day or less often
Daily or more often vs. 2Æ1 1Æ1, 4Æ3 85%. This broad range may indicate the differences in the
weekly or less often authors’ definition of a NCCLs or the ability of the
Weekly or more often 1Æ5 0Æ7, 3Æ2 examiners to detect the lesions. NCCLs were a common
vs. monthly or rarely finding in the patients sampled (prevalence rate 62Æ2%)
which is closer to the upper end of the prevalence rates
reported in the literature (1). Out of all the teeth
Occlusion examined, 13Æ1% had lesions which is similar to the
findings of Boric et al. (2) (16Æ6%) and Pegoraro et al. (16)
Of the patients examined, 19Æ7% exhibited canine
(17Æ23%). The teeth most frequently associated with
protected guidance on both right and left sides, while
lesions were the upper and lower premolars and upper
13Æ9% showed group function of both sides. Most
first molars which is similar to the findings of Bernhardt
patients (61Æ4%) had canine guidance on one side and
et al. (11); their data found that the upper and lower
group function on the opposite side. Patients with
premolars were at highest risk for developing lesions.
group function on one side or both sides had signifi-
Persons under 25 years of age had a significantly
cantly higher associations with NCCLs (on that side or
lower association with NCCLs than persons older than
both sides) when compared with those with canine
25 years (OR 4Æ6). Similarly, persons over 50 years of
guidance. There was a significant association between
age had a higher association with NCCLs than persons
the presence of NCCLs in patients who showed
<50 years of age (OR 4Æ1). When persons younger than
evidence of faceting. This is shown in Table 6.
65 were compared with persons over 65, we found a
higher association of having lesions in the over 65
Discussion group, but that association was lower than the other
two comparisons (OR 2Æ5). These data suggest that the
This paper examines the prevalence and severity of
older the patient, the greater the chance of having
NCCLs regardless of aetiology. The authors did not
lesions but that association begins to decline after age
attempt to classify the lesions according to aetiology or
25. The finding is at variance with Bernhardt et al. (11)
who found higher risks with increasing age though
Table 6. The strength of association between occlusal factors and their data did find a reduced risk in the 40–45 year age
non-carious cervical lesions expressed as odds ratio with 95% group. The size of the sample and the small number of
confidence intervals persons over 65 examined may account for the different
findings.
Patients with Patients with The NCCLs examined were mostly only into dentine
one or more no non-
(99Æ2%). Tertiary dentine was visible in only three
non-carious carious 95%
cervical cervical Odds confidence
(0Æ8%) of the teeth examined. This study looked only at
Variable lesions lesions ratio interval the number of un-restored lesions and therefore, the
data may actually under-represent the number of
Guidance
lesions into tertiary dentine as they may have been
R group function (R side) 76 (R Side) 25 1Æ3 ‘0Æ6, 2Æ9
R canine guidance (R side) 28 (R Side) 12 restored or the teeth may have been extracted.
L group function (L side) 77 (L side) 24 1Æ7 0Æ8, 3Æ8 Of all the lesions, 45% were sensitive to compressed
L canine guidance (L side) 26 (L side) 14 air. The high prevalence rates of patients with lesions
One or more 28 21 3Æ3 1Æ6, 6Æ6 and the finding that 45% of patients reported sensitiv-
facets present
ity to cold air suggests that NCCLs is a major problem in
No facets present 31 76
the population sampled. Dentists ought to be aware of

ª 2007 The Authors. Journal compilation ª 2007 Blackwell Publishing Ltd

 NON-CARIOUS CERVICAL LESIONS IN TRINIDAD
the high prevalence, be mindful that it is a frequent
cause of tooth sensitivity and be competent in the
diagnosis and management of the lesions when treating
Trinidadian patients.
The data on oral hygiene practices support the role of
toothbrush abrasion in the development of cervical
lesions and is in agreement with other authors (11).
Patients who reported frequent episodes of heartburn
or gastric reflux had a strong association of having
NCCLs which supports the role of intrinsic acids in the
development of these lesions. Extrinsic acids from citrus
fruits, effervescent soft drinks, yoghurt and effervescent
vitamin C tablets were shown as factors strongly
associated with the development of lesions. Further,
the more frequent the intake of some of these foods, the
greater the association with NCCLs. People who
engaged in swimming in chlorinated swimming pools
weekly or more often had a significantly higher
association with NCCLs than those who swam monthly
or less often. This again supports research which
demonstrates the role of erosion in the development
of NCCLs (6). Vegetarians were strongly associated with
NCCLs and this may warrant further investigation into
the nature of the vegetarian diet in Trinidad in terms of
what is contributing to the development of these lesions
though one study reported a link between dental
erosion and a lacto vegetarian diet (5).
Patients with faceting or with group function were
strongly associated with non-carious lesions. Similar
findings have been reported in the literature (11, 13,
16, 17) although many others failed to show a signi-
ficant relationship between the two (4). Also, patients
who reported bruxing, headaches, wearing an occlusal
splint, clicking joints or a history of broken restorations
at least once a year were strongly associated with the
presence of NCCLs. It is well known that parafunctional
habits can lead to patients having problems of fractured
restorations and muscular spasm which can cause
headaches and inter-articular disk ‘clicking’ of the
temperomandibular joints. The association of these
factors with NCCLs supports the theories regarding
the aetiological role of occlusal factors.
Some of the limitations of this study include the fact
that the sample was a convenient one and cannot
necessarily be representative of the entire population of
Trinidad. Further, treatment in the dental school is
subsidized and the population examined may be
skewed toward the lower income group. Also the data
obtained regarding medical conditions and diet were
ª 2007 The Authors. Journal compilation ª 2007 Blackwell Publishing Ltd
133
obtained from the patients themselves which would
contribute to errors of accuracy.
Conclusion
Non-carious cervical lesions are a common finding in
this Trinidadian sample especially in those over
25 years. Many of these lesions are symptomatic
causing patients to complain of sensitive teeth. The
teeth most commonly involved are upper and lower
premolars. A number of dietary factors, medical condi-
tions and occlusal factors seem to be associated with
NCCLs which lends support to the proposition that the
aetiology of these lesions is multifactorial.
References
1. Levitch LC, Bader JD, Shugars DA, Heymann HO. Non-
carious cervical lesions. J Dent. 1994;22:195–207.
2. Boric J, Anic I, Urek MM, Ferreri S. The prevalence of non-
carious cervical lesions in permanent dentition. J Oral
Rehabil. 2004;31:117–123.
3. Oginni AO, Olusile AO, Udoye CI. Non-carious cervical
lesions in a Nigerian population: abrasion or abfraction? Int
Dent J. 2003;53:275–279.
4. Bartlett DW, Shah P. A critical review of non-carious cervical
(wear) lesions and the role of abfraction, erosion and
abrasion. J Dent Res. 2006;85:306–312.
5. Linkosalo E, Markkanen H. Dental erosion in relation to
lactovegetarian diet. Scand J Dent Res. 1985;93:436–441.
6. Lussi A, Schaffner M. Progression of and risk factors for dental
erosion and wedge-shaped defects over a 6-year period. Caries
Res. 2000;34:182–187.
7. Lee WC, Eakle WS. Possible role of tensile stress in the
etiology of cervical erosive lesions of teeth. J Prosthet Dent.
1984;52:374–380.
8. Rees JS. A review of the biomechanics of abfraction. Eur
J Prosthodont Rest Dent. 2000;8:139–144.
9. Bader JD, McLure F, Scurria MS, Shugars DA, Heymann HO.
Case control study of non-carious cervical lesions. Commun
Dent Oral Epidemiol. 1996;24:186–191.
10. Mayhew RB, Jesse SA, Martin RE. Association of occlusal,
periodontal and dietary factors with the presence of non-
carious cervical lesions. Am J Dent. 1998;11:29–32.
11. Bernhardt O, Gesch D, Schwahn C, Mack F, Meyer G,
John U, Kocher T. Epidemiological evaluation of the multi-
factorial aetiology of abfractions. J Oral Rehabil. 2006;33:
17–25.
12. Bader JD, Shugars DA, Heymann HO, McClure F. How
dentists classified and treated non-carious cervical lesions.
J Am Dent Assoc. 1993;124:46–54.
13. Hand JS, Hunt RJ, Reinhardt JW. The prevalence and
treatment implications of cervical abrasion in the elderly.
Gerodontics. 1986;2:167–170.
134 W . A . J . S M I T H et al.

14. Miller N, Penaud J, Ambrosini P, Bisson-Boutelliez C, Brian-
con S. Analysis of etiological factors and periodontal condi-
tions involved with 309 abfractions. J Clin Periodontol.
2003;30:828–832.
15. Rafeek RN, Marchan S, Eder A, Smith WAJ. Tooth surface loss
in adult subjects attending a university dental clinic in
Trinidad. Int Dent J. 2006;56:181–186.
16. Pegoraro LF, Scolaro JM, Conti PC, Telles D, Peforaro TA.
Noncarious cervical lesions in adults: prevalence and occlusal
aspects. J Am Dent Asssoc. 2005;136:1694–1700.
17. Khan F, Young WG, Shahabi S, Daley TJ. Dental cervical
lesions associated with occlusal erosion and attrition. Aust
Dent J. 1999;44:176–186.
Correspondence: Dr William A. J. Smith, School of Dentistry, Faculty
of Medical Sciences, The University of the West Indies, Eric Williams
Medical Sciences Complex, Champs Fleurs, Trinidad, West Indies.
E-mail: wsmith@fms.uwi.tt

ª 2007 The Authors. Journal compilation ª 2007 Blackwell Publishing Ltd