A disadvantage of early operative intervention is the elimination of the potential for small incipient lesions to remineralize or arrest such that they do not need operative intervention. Resin-based composites, which most often are used to restore air-abraded teeth, require the use of a bonding agent for optimal results.^8,9 Currently used bonding agents and resin-based composites are subject to technical considerations such as moisture control, pooling of the low-viscosity agents and adequate light curing. This is problematic when treating posterior teeth in particular and can lead to a more technically challenging procedure. In addition, once restored, these teeth require continuing maintenance,^10,11 because of possible wear, microleakage or fracture of the restorative material.

Given that there are advantages and disadvantages associated with the early treatment of small, questionable carious lesions, we initiated a randomized controlled clinical study to investigate the merits of early treatment of these lesions. Although the application of pit and fissure sealants would be appropriate for teeth in all but the oldest patients in this study, we did not use sealants because the hypothesis to be investigated was the effect of early treatment of questionable carious lesions, not preventive measures to control the progression of these lesions. This report will review the two-year results of a proposed five-year study.

	PATIENTS, MATERIALS AND METHODS

TOP ABSTRACT PATIENTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
Before we recruited patients for this clinical study, the Institutional Review Board of the University of Michigan, Ann Arbor, examined and approved the protocol, as well as the consent and assent forms. The assent form, which is similar to a consent form, but written at a level to be understood by the youngest patients, was required to be signed by minors enrolled in the study. In addition, the consent form was signed by parents or guardians.

New and returning patients between the ages of 12 and 36 years were recruited from patients who received a routine dental examination at the University of Michigan School of Dentistry and had at least one questionable carious lesion in any of the pits and fissures of a posterior tooth. We made no distinction between pits or fissures on the buccal or lingual surfaces of the tooth. The diagnosis of "questionable" was made by dentists who were supervising dental students, but who were not part of the clinical study.

Patients were asked if they would be willing to participate in a clinical study to investigate the early treatment of very small carious lesions. If a patient indicated interest, he or she filled out a referral card that was given to the research coordinator (who was not a dentist) to schedule a baseline examination. We kept no records of patients who declined to participate.

We defined "questionable" as no frank caries detected by conventional examination (that is, softness, decalcification or cavitation at the base of a pit or fissure, or evidence of radiolucency seen on bitewing radiographs, which were available for all patients), but we were uncertain about whether caries was present when deep staining or explorer retention was observed. The vast majority of these questionable lesions were darkly stained pits and fissures.

We enrolled a maximum of three teeth from each patient to increase the variability of caries risk and to limit the number of teeth lost to follow-up if a patient missed a recall evaluation. Although not a requirement for selection, all enrolled teeth were in occlusion with natural dentition.

Patients were excluded from the study if they had five or more active carious lesions, which is an indication of high caries risk or significant oral health neglect that could influence the rate of caries development independently of the pit or fissure defect. We also excluded patients with a physical or mental handicap that would limit oral hygiene practices or predispose them to an unusually high caries risk. Although these exclusion criteria were placed in the protocol to limit the risk to patients enrolled in the study, no patients with questionable carious lesions were excluded owing to these restrictions.

Ninety-three patients with 223 questionable carious lesions were enrolled in the study. Each enrolled tooth was examined independently by two dentists who were selected from three of us (J.H., J.D., K.S.) based on our availability at the time of the baseline evaluations. We all used air from an air-water syringe, no. 13–14 explorers purchased and used only for this study, and x2.5 magnification to examine the study teeth. Each of the dentist examiners had a minimum of 25 years of clinical experience.

Each dentist probed the pits and fissures of each study tooth extensively. Each dentist evaluated the darkest color of the pit and fissure system, explorer retention in a pit or fissure and gingival health using the Löe and Silness Gingival Health Index.¹² One examiner scored the amount of plaque on each study tooth using the Simplified Oral Hygiene Index.¹³ Box 1 shows the evaluation criteria.

This DMFS score for each patient was linked to each treatment or control tooth in that patient. The mean DMFS scores for the treatment and control groups were calculated by averaging the linked scores for each tooth in the two groups. Erupted third molars were included in the DMFS score for the few patients who had them. None of the teeth with questionable carious lesions were partially erupted. After all baseline measurements were completed, each tooth was randomly assigned to an early treatment group or the control group based on a table of random numbers.¹⁵

Patients with teeth assigned to the early treatment group were told that although local anesthetic usually was not necessary when treating small lesions with air abrasion, the operator would be willing to administer it at any time if the patient desired. With the patient’s concurrence, all air-abrasion procedures were initiated without the injection of local anesthetic.

Air abrasion. After application of a rubber dam, the dentist air-abraded the questionable pits and fissures of the 113 teeth randomized into the early treatment group using 27-micrometer aluminum oxide powder starting at 80 pounds per square inch (5.8 kilograms per square centimeter). If the completed preparation extended into dentin—an evaluation made independently by the two dentists—an impression was taken to measure the lost tooth structure. Using the impression of the occlusal surface taken during the baseline examination to form the occlusal surface of the preparation impression, the dentist injected a quick-setting polyvinyl siloxane low-viscosity impression material into the preparation. The impression was weighed as a surrogate measure of the amount of tooth structure lost.

All teeth with preparations that extended into dentin and those with preparations that extended only into enamel were restored or sealed with a flowable light-cured resin-based composite (Tetric Flow, Ivoclar Vivadent Inc., Schaan, Liechtenstein). Every tooth in the treatment group received at least a prepared sealant (that is, for those preparations that were entirely within enamel). We believed that a lightly filled resin would be more appropriate for sealing these narrow enamel-only preparations.

If the preparation extended into dentin, the dentist placed a preventive resin restoration with the radiating fissures sealed, using air abrasion and sealing with a flowable resin-based composite. This was done according to the manufacturer’s instructions (that is, etching with phosphoric acid gel for 20 seconds, rinsing with water for 15 seconds and applying two light-cured coats of bonding agent [Syntac SC, Ivoclar Vivadent Inc.]). Finishing and polishing were performed using slow-speed burs and rubber points.

Recall examinations. Two of the four dentist authors re-examined the 113 treated teeth and 110 control teeth at six-month intervals. Although bitewing radiographs were obtained at yearly intervals as part of the patients’ continuing care at the University of Michigan, we did not find them to be effective in diagnosing the small occlusal carious lesions of interest in this study. This was most likely due to the limited size of the lesions and their varied location on the occlusal surface. At each recall examination, all study teeth were evaluated independently by two dentists for gingival health. The quality of the restorations was evaluated according to modified Ryge criteria,¹⁶ and the retention of sealants was assessed according to the following criteria: present, partially lost and completely lost (Box 2).

View this table: [in this window] [in a new window]   BOX 2 MODIFIED RYGE EVALUATION CRITERIA FOR THE EARLY TREATMENT GROUP AND TREATED CONTROL TEETH.*

The dental assistant or research coordinator recorded the independent scores for each criteria from the two dentists; if there was any disagreement, the dentists reviewed the written criteria and reached a consensus. The evaluators agreed 84 percent of the time before any review or discussion took place. (Agreement ranged from 65 percent for evaluation of pit and fissure feel to 91 percent for evaluation of anatomical form and presence of sealant.) All dentist authors, in pairs of two, were evaluators. Drs. Hamilton and Dennison prepared and placed all restorations, but did not necessarily evaluate all restorations or preparations at recall appointments. Dentist evaluators used x2.5 magnification and an explorer to evaluate caries into dentin and the elimination of carious tooth structure.

Table 1 shows the distribution of the 223 control and treated teeth among the 93 patients. The distribution of molars and premolars between the control and treatment groups is shown in Table 2.

Logistic regression with generalized estimating equations for clustered data was used to determine which baseline factors (that is, pit and fissure color, pit and fissure feel, baseline DMFS index, fluoride use history, age, sex and tooth type) were related to control teeth that were subsequently diagnosed as having occlusal caries, as well as to treated teeth that had caries extending into dentin. The weights of the preparation impressions—a surrogate measure of volume—of the early treated teeth and the control teeth subsequently diagnosed as having caries and treated were compared using a t test for independent samples. A P value of less than .05 was considered significant.

	RESULTS

TOP ABSTRACT PATIENTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
The dentists re-examined 89 control teeth at 24 months or longer (because patients were treated at staggered times). Nine (10 percent) of the teeth were diagnosed and treated for caries after one year; five additional teeth (6 percent) were diagnosed and treated in the second year. We found no significant effect of time on caries occurrence in the control group (P = .8470). This means that the probability of a questionable carious lesion in the control group being diagnosed with caries did not increase or decrease significantly during the two years. The probability of caries developing in teeth in the control group after two years was 16.1 percent (95 percent confidence interval, 9.3 percent to 26.3 percent). The number of control teeth diagnosed with caries during the first two years of the study was significantly less than the number of teeth with caries that extended into dentin in the treatment group at baseline (P < .001).

The mean weight of the impression material in preparations that extended into dentin (a surrogate measure of tooth structure volume) in the early treatment group was 0.0260 grams, compared with 0.0281 g in the control group. This was not a statistically significant difference (P = .390). The only significant baseline predictor of a control tooth being diagnosed with caries was pit and fissure feel evaluated with an explorer (P = .0149). This means that the more retentive the explorer was in the pit or fissure, the more likely the control tooth would be diagnosed as having caries and treated during the next 24 months.

Table 3 shows the results of clinical evaluations of preventive resin restorations and sealants placed at baseline using modified Ryge criteria¹⁶ at six months, 12 months, 18 months and 24 months. During the second year of the study, no restorations required re-treatment compared with two restorations needing re-treatment during the first year of the study. Table 4 (page 1649) shows the results of the evaluations of gingival health and the plaque index for all teeth and pit and fissure color and pit and fissure feel of control teeth at six months, 12 months, 18 months and 24 months. Figure 1 (page 1650) shows the 24-month appearance of a tooth in the early treatment group. Figure 2 (page 1650) shows the 24-month appearance of a tooth in the control group.

View larger version (97K): [in this window] [in a new window]   Figure 1. Preventive resin restoration in the mandibular first molar at 24-month recall examination.

View larger version (41K): [in this window] [in a new window]   Figure 2. Control tooth (mandibular second molar) at 24-month recall examination.

	DISCUSSION

TOP ABSTRACT PATIENTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

At baseline, we discovered that caries had progressed into dentin in 50 (44 percent) of the 113 teeth with questionable carious lesions randomized into the early treatment group. After two years, only 14 teeth randomized into the control group were diagnosed with caries. The criteria we used for diagnosing caries throughout the study was softness at the base of a pit or fissure, decalcification associated with a pit or fissure, or cavitation.

Probed teeth. We probed control teeth extensively with an explorer. This leads to two concerns: the possible transmission of cariogenic bacteria from one tooth to the next, and damage to decalcified enamel and increased demineralization.^18,19 If these events took place, we would expect that a greater number of control teeth would be diagnosed and treated for caries than would otherwise be the case. However, this did not happen during the 24 months of this study.

Cariologists’ estimate. Before the study began, we consulted with two cariologists to obtain an estimate of the percentage of control teeth with questionable carious lesions that would progress to unquestionable caries each year of the study. Although no previous study examined only questionable carious lesions, the cariologists’ estimate was that 25 percent of the teeth would become carious each year. Since 44 percent of the teeth randomized to the early treatment group had clinically demonstrated caries that had progressed into dentin when treated at baseline, we expected to find more than 25 percent of the control teeth exhibiting caries during each of the first two years.

One possible explanation for the low number of control teeth exhibiting caries each year is that the rate of progression of caries in this population is lower than expected. Other possible explanations are that the caries had arrested or remineralization had occurred, although these possibilities run counter to current thinking about progression of pit and fissure caries that has extended into dentin. Current explanations may have to be reconsidered if the caries rate in the control group remains low.

DMFS scores. Another possible explanation for a lower-than-expected caries rate might be related to the caries experience or susceptibility of the enrolled subjects. One measure of dental caries experience or prevalence is the number of DMFS present. The DMFS score for enrolled patients ranged from 0 to 58, with a mean of 8.42 and a standard deviation of 7.47. Figure 3 (page 1650) illustrates the distribution of DMFS scores and their association with study teeth. It is interesting to note that four teeth in the study were in patients with a DMFS score of 0, while one patient (one tooth) had a DMFS score of 58. Thus the caries prevalence varied within the study.

View larger version (90K): [in this window] [in a new window]   Figure 3. Caries prevalence as indicated by decayed, missing and filled surfaces, or DMFS, scores of patients associated with study teeth.

As noted in the 12-month results,²⁰ demographic and baseline evaluation variables associated with teeth in the early treatment group were analyzed using logistic regressions with generalizing equations. DMFS was not significantly related to caries extending into dentin in the early treatment group, but was borderline (P = .056). The surprising fact is that the higher the DMFS score was at baseline, the less likely that caries had extended into dentin in teeth randomized into the treatment group. One possible explanation is that the most susceptible surfaces had already been restored, leaving only those surfaces that were more resistant to caries.

In addition, this could be construed as the lower one’s DMFS score, the more likely that caries had extended into dentin and that the control group, which had a lower overall DMFS, would have more caries into dentin than the early treatment group, which had a higher DMFS. The fact is, however, that despite the lower mean DMFS score in the control group and our use of aggressive probing, the caries rate was still lower than expected in the control group.

When comparing the size of the preparations in the early treatment and control groups, we found it surprising how small these preparations were and how small the difference was between the two groups. (The mean weights of the impressions from the two groups differed by only 0.0021 g, or 1.6 cubic millimeters in size.)

Darkly stained pits and fissures often have been associated with caries. In this study, these pits and fissures were divided into five categories ranging from tooth colored (nearly white) to black. This color range can be considered a continuous scale, and it was clear to the evaluators that in certain cases, there would be difficulty distinguishing a difference in adjacent categories, such as between tooth color and light yellow, between light brown and dark brown, and between dark brown and black. Consequently, the level of agreement (84 percent) between the evaluators was not unexpected. When we reduced the color scale to two categories (no stain [tooth color and yellow/orange] and stained [light brown, dark brown and black]), the agreement between evaluators was an acceptable 88 percent.

Use of dental explorer. Using a dental explorer to probe the pits and fissures of teeth has often been an aid in diagnosing carious lesions²¹; however, many cariologists do not recommend the use of aggressive probing.^18,19 Although specific explorers were purchased and used in this clinical study, we found differences between similar explorers. We also noted that the explorers were used with different pressures by different examiners. In retrospect, we should have developed pressure criteria.

Moreover, it was clear to all examiners that physical changes to the pits and fissures occurred as a result of probing that made calibration impossible. In many cases, the pit or fissure became less retentive the more the tooth was probed. Given these problems, it is interesting to note that we found a strong relationship between pit and fissure feel and caries penetrating into dentin (P = .0149). The greater the retention of the explorer at the baseline examination, the more likely the control tooth would be diagnosed and treated for caries during the next 24 months.

Although patients were excluded from the study if they had five or more active carious lesions, only one patient had an extensively cavitated lesion on enrollment. In retrospect, this might have been expected, because if a patient has many grossly carious teeth, it could indicate a high caries rate. We would expect that teeth with questionable carious lesions would not stay questionable for long in a patient with a high caries rate.

Caries-detecting dyes. Before this study began, we decided not to use any caries-detecting dye to aid in locating or removing carious dentin. This decision was based on two factors:

– A review of the published literature indicated diverse opinions regarding the effectiveness of caries-detecting dyes.

– We wanted to use procedures that were commonly practiced by general dentists.

In a review article on the efficacy of caries-detecting dyes, McComb²² noted that with proper dental care, they were not necessary.

None of the patients requested local anesthetic during treatment of the five control teeth in the second year of the study, compared with one patient in the first year. No patient reported any preoperative or postoperative sensitivity associated with any study tooth. Both of these findings suggest that the questionable carious lesions in the control group are not progressing quickly.

This study did not measure the amount of tooth structure lost as a result of early treatment of the 63 teeth in which the questionable carious lesions were limited to enamel. We also did not predict the costs related to the lifetime maintenance these prepared sealants require. The lost tooth structure and lifetime maintenance costs need to be weighed against the benefits of treating caries at an early stage.

	CONCLUSION

TOP ABSTRACT PATIENTS, MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
The results of this study show that after two years, tooth structure was not conserved as a result of early treatment of questionable carious lesions in posterior teeth. The degree of explorer retention at baseline was significantly associated with caries being diagnosed later in control teeth. The preventive resin restorations and sealants placed in the early treatment group were performing well after two years.

View larger version (136K): [in this window] [in a new window]   Dr. Hamilton is a clinical associate professor, Department of Cariology, Restorative Sciences and Endodontics, the University of Michigan School of Dentistry, 1011 N. University, Ann Arbor, Mich. 48109-1078, e-mail "jchamilt{at}umich.edu". Address reprint requests to Dr. Hamilton.

View larger version (131K): [in this window] [in a new window]   Dr. Dennison is a professor, Department of Cariology, Restorative Sciences and Endodontics, the University of Michigan School of Dentistry, Ann Arbor.

View larger version (139K): [in this window] [in a new window]   Dr. Stoffers is an assistant professor, Department of Cariology, Restorative Sciences and Endodontics, the University of Michigan School of Dentistry, Ann Arbor.

View larger version (129K): [in this window] [in a new window]   Dr. Gregory is an adjunct professor of dentistry, Department of Cariology, Restorative Sciences and Endodontics, the University of Michigan School of Dentistry, Ann Arbor.