Diagnosis of approximal caries
The issues to be considered by the clinician with respect to caries of the approximal surfaces are similar to those considered at other sites: Is the surface sound, or is there a lesion? If so, how advanced is the lesion¾involvement of enamel only, enamel and dentinal involvement, or pulpal exposure? Finally, is there cavitation?
Diagnostic methods
Meticulous visual examination
In the thin anterior teeth, both noncavitated and cavitated approximal lesions are readily detectable by meticulous clinical visual examination. The diagnostic potential is improved by fiber-optic transillumination.
Temporary tooth separation
The approximal surfaces of the posterior teeth cannot usually be viewed directly, and temporary elective tooth separation has been recommended as an inexpensive, reversible method of clinical examination (Pitts and Longbottom, 1987) (see Figs 190, 191, and 192). The main disadvantage is the need for two appointments.
Supplemented by impressions, this method offers the highest sensitivity and specificity both for enamel (D1) and dentin (D3) lesions (see Fig 200). It also allows the clinician to determine whether the lesion is noncavitated or cavitated and to estimate the size and depth of the cavity. The impression itself, or a cast replica, can be stored for future reference, to allow monitoring of progression of cavitation.
Pitts and Rimmer (1992) applied this method in a study of more than 200 schoolchildren and correlated the findings with the appearance of the lesions on bitewing radiographs: None of the radiographic lesions in the outer half of the enamel exhibited cavitation, 10% of the lesions in the inner half of the enamel and 40% of lesions penetrating to half the depth of the dentin were cavitated, and all the lesions involving more than half the depth of the dentin exhibited cavitation. However, no data were provided on cavity depth, ie, the numbers of cavities located in the enamel
and in the dentin.
Lunder and von der Fehr (1996) applied the same technique to investigate the relationship between cavitation and carious activity, by comparing the prevalence of cavitation in radiographic enamel lesions extending to the dentinoenamel junction (D2) and into the dentin (D3) in 17 to 18 year olds with high or moderate carious activity. Cavitation of D2 lesions was very rare in moderately caries-active subjects, while there was usually cavitation of both D2 and D3 lesions in highly caries active subjects. However, the depth of the cavities¾whether involving only superficial enamel or extending further into the enamel¾was not evaluated.
All noncavitated approximal lesions, not only in enamel (D1 and D2), but also in
dentin (D3), can be arrested by intensified, targeted plaque control and use of fluoride.
It is unlikely that cavitated lesions will be arrested, because plaque will be retained in
the inaccessible cavity. The risk of overtreatment by operative intervention can be
minimized by applying the temporary tooth separation method to differentiate
between noncavitated and cavitated approximal lesions. The method warrants more
widespread application in developed countries with well-organized dental care and
declining caries prevalence.
Radiographs
To date, the bitewing radiograph has been the standard diagnostic method for
detecting approximal lesions, and its value is beyond dispute. Radiographs usually
disclose 50% more (30% to 69%) small approximal lesions than does clinical
examination alone. As discussed earlier, there are three possible outcomes for such
small approximal lesions: progression, arrest, or regression (Meja`re et al, 1999).
In Swedish preschool children (4 to 6 year olds), Moberg-Skold et al (1997) detected
3.8 approximal carious lesions per child by using bitewing radiographs, compared to
only 2.0 lesions per child by meticulous clinical visual examination. However, in
schoolchildren with declining caries prevalence, the efficiency of bitewing
radiographs in diagnosing approximal lesions warrants critical reassessment. De Vries
et al (1990) concluded that, for diagnosis of approximal caries, there would not be
substantial loss of information without radiographs in children younger than 12 years,
but that radiographs should be included in examinations of older children.
The frequency of monitoring of these age groups with periodic bitewing radiographs
is related to the risk category: annually for high-risk patients and every 2 to 4 years
for low-risk patients (the risk category should be reassessed at each visit before
radiographs are taken). In children at the mixed dentition stage, a 1- to 2-year interval
is recommended. In Western industrial societies, progression of approximal enamel
lesions to the stage at which there is radiographic evidence of extension into dentin is
now only gradual (3 to 9 years), allowing ample time to monitor lesion progression,
arrest, or regression (see Meja`re et al, 1999).
Bitewing radiographs used for this purpose must be standardized with respect to
positioning, exposure, and development of the films, and overlapping of the teeth
must be avoided. Film-holding beam-aiming devices are useful for standardizing the
position of the film and the angulation of the beam. In addition, films should be read
with proper lighting and magnification.
The use of ionizing radiation is not without hazard, and all efforts must be made to
minimize radiation exposure, consistent with achieving a diagnostically useful image.
The radiation dose can be reduced by collimation of the x-ray beam, ensuring that the
timer is functioning accurately, avoiding technical errors and attendant retakes, using
fast E-speed film, and protecting the patient with a lead apron and a thyroid collar.
For systematic longitudinal monitoring of approximal caries, several scores or indices
have been proposed. The D1 to D4 scale was described earlier in this chapter. The
following three-point scale for radiographically detectable lesions was introduced
some years ago by the Swedish Board of Health and Welfare:
· D1: lesion in the outer half of the enamel
· D2: lesion in the inner half of the enamel
· D3: lesion in dentin
Figure 215 illustrates the five-point Norwegian system for scoring radiographs
(Espelid and Tveit, 1984):
· Grade 1: lesion in the outer half of the enamel
· Grade 2: lesion in the inner half of the enamel but not into the dentin
· Grade 3: lesion into the outer third of the dentin
· Grade 4: lesion into the middle third of the dentin
· Grade 5: lesion into the inner third of the dentin
Treatment decisions: Prevention versus extension
An unresolved question is not the detection of approximal caries, but at what stage
restorative treatment is indicated. Some dentists believe that all lesions should be
restored, irrespective of extent. In regions where carious activity remains relatively
high, operative intervention for an approximal lesion with a radiolucency confined to
enamel is not unusual; elsewhere, it is considered unethical to restore such early
lesions. To restore lesions that are not active and progressively extending through the
enamel is tantamount to malpractice. A majority of dentists consider that an
appropriate end point for nonoperative intervention is radiographic evidence that the
lesion has extended into the dentin. However, provided that there is no cavitation in
the dentin, the dentinal tubules have not been invaded by microorganisms, and the
lesion can still be arrested. Operative intervention for noncavitated lesions of dentin
should therefore also be regarded as malpractice.
The dentist should also be aware that restoration of the approximal surfaces is not
without its hazards. One recent study reported a high frequency of iatrogenic damage
(about 66%) associated with cavity preparation for restoration of approximal lesions
(Qvist et al, 1992). A so-called proxitector should be used to protect the neighboring
approximal surface when rotating instruments are used. Such a surface may have a
noncavitated lesion that could be arrested.
In teenagers and young adults, the cervical margins of approximal restorations are
frequently placed subgingivally, because the interproximal “space” beneath the
contact area is completely occupied by the buccal and lingual papillae and the col.
The cervical margins of most such restorations are inadequately finished, some have
overhangs, and composite materials are particularly plaque retentive. By enhancing
plaque retention, these factors contribute to increased risk for recurrent caries and
development of periodontitis.
In determining whether invasive therapy is indicated, the clinician must address the
following questions:
1. How rapidly is the lesion progressing?
2. What is the size or depth of the enamel or dentin lesion (see the Norwegian fivegrade
scale)?
3. Is the lesion noncavitated or cavitated? This may be determined by temporary
separation if no cavity is found by meticulous clinical examination.
4. What is the patient’s predicted caries risk and risk profile?
5. Why have the patient’s self-care and the professional preventive measures failed to
prevent development of the lesion?
6. How can combined preventive efforts be improved to arrest the lesion?
7. How much time should elapse before the outcome of preventive efforts is
evaluated?
The second and third points in this list were evaluated by Bille and Thylstrup (1982)
in a study of 8- to 15-year-old children. The subjects were examined clinically and
radiographically before undergoing regular dental care, including restorations.
Approximal lesions were scored on the radiographs according to the following system
(Fig 216), which is similar to the diagnostic scales proposed by Moller and Poulsen
(1973), and Grondahl et al (1977):
· Score 0: enamel with no radiographic changes
· Score 1: radiographic changes in enamel
· Score 2: radiolucent lesion reaching the dentinoenamel junction
· Score 3: radiolucent lesion penetrating approximately halfway through dentin
· Score 4: radiolucent lesion close to the pulp
During cavity preparation, drilling was discontinued when the maximal extent of the
lesion could be seen on the base of the approximal box, cervical to the interproximal
contact area. With the help of an intraoral mirror and probe in normal clinical lighting,
the tissue changes observed at the base of the approximal box were classified
according to the six-point clinical scoring system (see Fig 216). This became the gold
standard for diagnosis of the approximal lesions:
· Scores 1 and 2: progressive changes in the enamel
· Score 3: changes in dentin, without cavitation in the enamel
· Scores 4 and 5: changes in dentin and progressive cavitation in the enamel (ie, at
this stage no bacterial invasion of the dentinal tubules has occurred, and there is no
indication for operative intervention)
· Score 6: cavitation involving dentin (possible indication for “drilling, filling, and
billing”)
Figure 217 shows a bitewing radiograph from the study, showing radiolucent enamel
lesions, radiographic score 1, on the distal surface of the mandibular right first molar
and the mesial surface of the mandibular right second molar. The radiolucency of an
approximal enamel lesion represents the sum of the loss of mineral from every single
enamel prism: At this stage, there has been no destruction of prisms. A cavity
prepared by drilling from the occlusal surfaces to the depth of the lesions observed on
the radiographs disclosed that these are enamel lesions without cavitation (clinical
score 1), limited to the distolingual surface of the first molar and the mesiolingual
surface of the second molar. Note the location of the lesions and the close proximity
to the inflamed gingival margin and the etiologic factor: the supragingival and
subgingival plaque (Fig 218).
Such enamel lesions should be arrested by use of a fluoridated wooden toothpick
twice a day from a lingual direction, supplemented at needs-related intervals by
PMTC with reciprocating pointed triangular tips and application of fluoride varnish.
As discussed earlier, in vivo studies by von der Fehr et al (1970) and Holmen et al
(1987) (see Figs 159, 160, 161, and 162) and in vitro studies by Silverstone, (1973)
(see Figs 157 and 158) have shown that experimentally induced enamel lesions can
successfully be arrested.
If the approximal surfaces can be maintained as sound, or with noncavitated lesions,
through the period of secondary maturation, the risk for future cavitation is
diminished (Kotsanos and Darling, 1991).
The relationship between radiographic and clinical scores is shown as a crosstabulation
in Fig 216. Of 158 lesions, radiolucencies penetrating approximately
halfway through the dentin (radiographic score 3) were found in only 58, and none of
these exhibited clinical cavitation into the dentin (clinical score 6). Only two of nine
lesions with a radiographic score of 4 had a clinical score of 6 (Bille and Thylstrup,
1982). Other studies have shown similar results (Mejare and Malmgren, 1986; Pitts
and Rimmer, 1992).
These studies confirm the importance of not interpreting radiographic findings in
isolation, to ensure that appropriate treatment decisions are made about approximal
lesions. Rather, radiographic findings must be assessed in the context of data from
other sources: temporary tooth separation and meticulous clinical visual examination
as well as careful consideration of each of the seven questions listed earlier.
Of the 158 carious lesions scheduled for restoration, cavity preparation disclosed that
66% were without macroscopic cavitation. The changes observed by direct clinical
inspection of the tissues during cavity preparation correlated poorly with accepted
standardized radiographic criteria. Thus cavitation was observed in only 20% of
lesions with radiolucencies extending to the dentinoenamel junction and in 50% of
lesions with radiolucencies involving the dentin, and all cavities were confined to the
enamel.
Bille and Thylstrup (1982) concluded that: “Assuming a macroscopical cavity into the
dentin to be indicative of restorative treatment, the present results indicate that a more
individualized treatment decision strategy than hitherto is warranted in populations
attending comprehensive dental health care.” Although this study was published in
1982, it seems to have had little impact on treatment decision making as taught in
undergraduate dental education or as practiced by most general practitioners.
In a Swedish study, dental hygienists and general practitioners examined approximal
lesions on radiographs of 100 extracted teeth and conducted meticulous clinical visual
examinations of more than 200 patients. No statistically significant differences were
found between dental hygienists and general practitioners in the inaccuracy in
detecting and recording lesions. It was concluded that, by undergoing examination by
a dental hygienist instead of a general practitioner, no patient with restorative
treatment need would have been overlooked, and a more accurate nonrestorative
(preventive) treatment need might have been addressed (Ohrn et al, 1996).
In a later large-scale study, Thylstrup et al (1986) reaffirmed their earlier findings
(Bille and Thylstrup, 1982) by using the same diagnostic scoring systems. At the time
of operative treatment, 263 Danish dentists recorded clinical tissue changes in 1,080
approximal lesions. The findings were related to type of tooth, age of the patient, and
information available from radiographs. The tissue changes observed varied
considerably, ranging from barely discernible white-spot lesions to frank cavitation.
This variation occurred irrespective of information available from radiographs. The
most frequently observed stage of lesion formation was a stage preceding cavitation.
Of 78 lesions with a radiographic score of 3, true cavitation was observed in fewer
than 10%. Of primary molars, only three exhibited cavitation into the dentin.
Cavitation into dentin was observed in only three of 330 lesions with radiographic
scores of 3, and in 27 of 102 lesions with radiographic scores of 4. With the current
potential for arresting caries through noninvasive methods, the findings indicate the
need to review the rationale for operative treatment of approximal lesions.
In another in vivo study, Meja`re et al (1985) examined 63 teenagers who were to
have premolars extracted for orthodontic indications. Comparison between direct
clinical examination after extraction and radiographs showed that noncavitated lesions
were the most frequent finding on the tooth surfaces examined (305 noncavitated
versus 28 cavitated). Of the noncavitated lesions, about 66.6% (n = 203) were not
detected by the bitewing radiographs. When the findings from direct inspection were
compared with those from clinical examination with an explorer, the accuracy of
diagnosing the absence of cavitation was 98.0% for incipient lesions and 100.0% for
sound tooth surfaces. Only 28.5% of the cavitated surfaces observed by direct
inspection were identified by probing.
In a supplementary study by Meja`re and Malmgren (1986), clinical tissue changes
were recorded during restorative treatment of approximal lesions in young premolars
and molars. Sixty approximal surfaces with radiographic radiolucencies in the inner
half of the enamel or the outer half of the dentin were treated. The extent and
character of the tissue changes were documented by magnification of photographs
taken during cavity preparation. The maximum extent of each lesion was correlated to
the extent of the radiographically observed lesion.
Cavitation was observed in 78% of the lesions with radiolucencies in the outer half of
the dentin, which is considerably higher than was found in the previously described
studies by Bille and Thylstrup (1982), Thylstrup et al (1986), and Pitts and Rimmer
(1992). However, these cavities were mainly limited to the enamel. In all cases, there
was discoloration of the enamel; the dentin was soft and discolored in 83% of the
lesions. In 12%, the tooth substance was severely damaged. Figure 219 shows
examples of the lesions. With the preventive materials and methods available today,
in theory the lesions that had not progressed to cavitation of dentin could have been
arrested. However, the plaque-retentive cavities, although limited to the enamel, had a
less favorable prognosis for caries arrest.
These studies confirm the importance of supplementing radiographic examination of
approximal carious lesions with clinical inspection prior to treatment decisions. For
ethical reasons, in the study by Pitts and Rimmer (1992), visual inspection was
achieved by using orthodontic elastometric separators for temporary tooth separation.
Bjarnason (1996) used a similar technique for temporary separation, not only for
clinical inspection, but also for minimally invasive operative intervention. The
restorative technique combined an external approximal composite filling, placed
under rubber dam isolation, and so-called partial tunneling.
Atraumatic restorative technique
Based on the current knowledge the following example illustrates recommended
procedures for noncavitated approximal lesions of enamel and dentin and for lesions
with limited cavitation: In a new patient, bitewing radiographs reveal one or more
posterior approximal lesions in the outer half of the dentin. Meticulous clinical visual
examination discloses no cavitation.
At the same or the following appointment, plaque is mechanically removed by PMTC,
in which reciprocating tips (Eva-Profin) are used on the affected approximal areas
(Figs 220 and 221). A slow-release chemical plaque control agent (Cervitec 1%
chlorhexidine-thymol varnish) is applied to these surfaces to destroy the remaining
cariogenic microorganisms (Fig 222). An orthodontic elastometric separator is then
inserted for temporary elective tooth separation to allow clinical visual inspection
after 5 days (Figs 223 and 224).
At the second appointment, after PMTC, the approximal surfaces are inspected
directly. Elastomeric impressions may be taken, to allow replication as records for
future reference. Lesions showing no cavitation are coated with chlorhexidine and
fluoride varnish (Cervitec + Fluor Protector) to arrest and seal the outer micropore
surface of the lesions (see Fig 223). If a limited cavity is diagnosed, it is mechanically
cleaned with a small ball-shaped finishing bur and filled with light-cured glassionomer
cement, placed under pressure from a translucent matrix band. After they are
finished with an extrathin tungsten-coated reciprocating tip, these surfaces are coated
with chlorhexidine and fluoride varnish to arrest and “seal” the surface of the
surrounding, noncavitated enamel lesions (see Fig 224).
