INTRODUCTION

It is well known that orthodontic patients are more prone to development of white spot lesions (WSL), a finding that is mainly attributed to plaque accumulation and food retention around the brackets. WSL may progress to cavitation or maintain as an eternal “scar” on the buccal enamel.1 Thus, it would be imperative to be able to detect those patients who are at higher risk for caries and WSL formation before the orthodontic treatment in order to implement appropriate and evidence-based preventive strategies. A multifactorial, computer-based caries risk assessment model, Cariogram,2 has been used and validated in schoolchildren and adult patients.3–6 Cariogram is an interactive PC program for caries risk evaluation that graphically shows a calculated overall caries risk for future caries after a weighted analysis of different etiologic variables of the patient.2 The Cariogram program is based on the input of several variables taken from a patient interview or questionnaire and the clinical examination as well as saliva testing.

Cariogram has previously been applied in orthodontic practice for detecting patients at risk for WSL development in order to implement appropriate preventive measures.7–9 In these previous studies, all 10 variables were employed according to the original manual of the Cariogram. Saliva tests are, however, rarely conducted during routine orthodontic treatment in Greece as a result of the extra costs, special equipment, and trained personnel required for such testing. In a previous study10 conducted in schoolchildren it was found that the accuracy of the Cariogram model was impaired when salivary mutans streptococci (MS), secretion rate, and buffer capacity were omitted. The purpose of this clinical study was therefore to compare the caries risk profiles according to Cariogram performed with and without some salivary variables in Greek adolescents undergoing treatment with fixed orthodontic appliances.

MATERIALS AND METHODS

Caries Risk Category

Table 1. Definition of Cariogram Scores and Percentage Distribution of Variables in the Study Group

View Fullscreen

The data were calculated in four ways: (1) Cariogram with nine variables, (2) Cariogram with SBC excluded, (3) Cariogram without salivary MS, and (4) Cariogram with both SBC and MS excluded. The patients were then allocated into three caries risk categories according to their chance of avoiding caries in the near future, as follows: 0–40% chance to avoid caries  =  high caries risk, 41–60% chance to avoid caries  =  medium caries risk, and 61–100% chance to avoid caries  =  low caries risk.

RESULTS

The distribution of Cariogram scores in the study group is shown in Table 1, and the mean values of some key variables are presented in Table 2. According to the information collected from the questionnaire, 58% of the patients were brushing their teeth twice daily with fluoridated toothpaste but more than half did not benefit from any additional fluoride measures. The SBC was uncompromised in the majority of cases (87%). High levels of salivary lactobacilli and/or MS were found in 40–50% of all patients, although almost 40% were found to be free from obvious decay (DMFT  =  0), and the mean DMFT value was 2.3. Furthermore, 28 of the 90 patients (31%) displayed at least one WSL adjacent to a bracket base.

Table 2. Age, Clinical Variables (DMFT, Plaque, Gingivitis, Number of white spot lesions [WSL] per Patient), and the 9-Item Cariogram Outcome

View Fullscreen

According to the nine-item Cariogram model, 62% displayed high caries risk, 13% displayed medium caries risk, and 24% displayed low caries risk (Table 3). When the SBC was omitted, 96% of the patients remained in the same risk category. Without the salivary MS counts, 23% of the patients changed their risk categories, most of them moving to a lower risk level. The distribution of the risk categories was, however, not significantly altered when compared with the nine-item Cariogram (P  =  .07). When the Cariogram was reduced by both SBC and MS counts, 76% of the patients were unchanged with respect to their risk categories.

Table 3. Comparison of Caries Risk Categories Obtained With the 9-Item Cariogram and Without Salivary Buffer Capacity (SBC) and/or Mutans Streptococci (MS)^a

View Fullscreen

DISCUSSION

The Cariogram risk assessment model has previously been applied to various age groups as well as to orthodontic patients. In some studies,16,17 certain age-related adjustments have been suggested, but we followed the “original” manual, with two exceptions. First, the salivary secretion rate was not measured, since the sample consisted of healthy teenagers and none of them displayed any clinical sign of impairment in terms of salivary flow rate. Second, the variable “clinical judgment,” which allows the clinician to add a subjective opinion of the patient, was automatically given a higher value when presence of any early sign of WSL was scored. The clinical classification system by Gorelick et al.,14 rather than the newer ICDAS-II Index,18 was used for the clinical scoring of WSL because it was more commonly used in the literature and allowed comparisons with previous findings. The chair-side tests were all considered accurate, reliable, and easy to use.19

The main finding of this study was that on the group level, no very dramatic differences in the distribution of risk categories appeared when the nine-item Cariogram was reduced. The low risk category remained around 20%, while the high risk category ranged from 54% to 63%. The minute impact of the SBC was expected in light of the high prevalence of noncompromised levels, while the stronger influence of the MS counts was more in line with previous findings.10 On the individual level, however, the risk category was changed for better or for worse, with the medium category being most volatile. For example, omitting the salivary MS counts, the 12 patients who shifted to a lower risk category were all caries free, had low plaque levels, participated in frequent dental visits, and received frequent fluoride application. On the other hand, the nine patients who shifted to the higher risk category had all caries, high plaque levels, rare dental visits, and infrequent fluoride application. These findings illustrate the value of the Cariogram model as a didactic help for individually targeted motivational interviews and counseling.

The clinical implication of our findings was clear cut: by omitting the salivary MS counts, a higher proportion of “medium-risk” patients was obtained, which eventually would not lead to a costly preventive “overtreatment.” Nevertheless, when it comes to prevention, overtreatment is definitely to be preferred over undertreatment, at least as long as serious side effects are uncommon and the costs are reasonable. It should also be stressed that WSL development during orthodontic treatment with fixed appliances is an underestimated problem; the incidence is high, and such lesions have very limited ability to improve after appliance removal.20 Consequently, orthodontists could safely be advocated to adopt the Cariogram model, even without the SBC and MS enumeration. The model is simple to use and easily understood by laypeople.7 In addition, the use of a formal caries risk assessment model adds structure, consistency, and quality to the risk assessment process and subsequent preventive work and is today regarded as “best clinical practice.”21 A more philosophical question would be whether all patients with fixed appliances should be regarded as being at risk as a result of the jeopardized oral cleaning. In this study, depending on the general incidence of WSL in the population, the answer is over 30%. Generally speaking, in populations with a low incidence the need for an individual risk assessment is more urgent than in those with a high incidence; in the latter, a group/community-based preventive approach is preferred.22

The mean Cariogram value of “actual chance of avoiding new cavities” was 36%. This was within the range of previous studies (28–75%), albeit at the lower end.8,9 This large variation may likely be due to the different study designs. The caries risk of the orthodontic patients was previously assessed at the end7 of their orthodontic treatment, while we estimated the risk within 18 months after the application of the appliances. Furthermore, no radiographs were utilized in the present study, in contrast to previous projects.7–9

Unfortunately, the present cross-sectional study design did not allow a validation of the risk assessment made against the true incidence of WSL at the time of debonding. Therefore, the accuracy and the predictive values of using the Cariogram in fixed orthodontic patients are still to be established. However, a previous study10 in schoolchildren showed clearly that the predictive power of the Cariogram was significantly impaired without performing the salivary tests. As concluded by several systematic reviews,19 the strongest single predictor for more caries is previous caries experience. Transferred to the orthodontic context, a close examination of recent bitewing radiographs before onset of appliances is recommended. A study by Fornell and Twetman23 strongly associated such findings with WSL development during treatment with fixed appliances.

CONCLUSIONS

• On the group level, the caries risk categories in patients with fixed orthodontic appliances were not significantly altered, when the Cariogram model without MS enumeration was employed, although a higher proportion exhibited a medium caries risk.

• Orthodontists can safely be advocated to adopt the model, even without saliva testing.