INTRODUCTION
Pits and fissures (P&F) of enamel surfaces in deciduous and permanent teeth are susceptible areas in which biofilm, food and bacterial debris can accumulate, leading to the development of acid-producing plaque resulting in caries 1 . The carbohydrates from food will be broken down, leading to demineralization of enamel. The extent of P&F is difficult to measure during a clinical examination. There are many investigations reporting different protocols for cleaning P&F 2 and for sealant application 3-5 .
In a study published in 2007, Cruvivnel et al 6 suggested that the depth of P&F is the most critical factor in terms of possible pulpal involvement. They showed that in the permanent dentition, the deepest part of the P&F is frequently located near or in direct contact with the enamel-dentine junction. They also hypothesized that in these cases, resident bacteria may leak into the pulp via dentinal tubules, which in turn may explain why some patients may complain of pain in “clinically healthy molars” 7-9 . The literature related to P&F depth and caries incidence in permanent molars is scarce 10 . The aim of this descriptive ex vivo assay was to evaluate qualitatively the depth of penetration of P&F in the enamel of human permanent third molars. The null hypothesis tested was that the enamel-dentine junction is not compromised by the depth of the P&F.
MATERIAL AND METHODS
The experimental protocol of the present study was authorized by the Institutional Research Ethics Committee of the Argentine Dental Association (Approval Code # 2019/0118-AOA). For this study, fifty (n=50) extracted human third molars stored at 4 °C in 2% thymol in normal saline were used. Inclusion criteria were being free of caries, restorations, pigmentations or other morphological alterations of the enamel, and having a centralized P&F system on the occlusal surface with at least one central fossa. The absence of clinically undetectable caries was confirmed with a laser fluorescence device (Diagnodent; Kavo, Biberach, Germany) as described by Lussi et al 9 .
After removal of gross debris attached to the roots, the occlusal surface of each tooth was cleaned for 30 seconds with a pumice/water slurry in a rubber cup at low-speed. The samples were then rinsed for 20 seconds with an air-water spray and dried with oilfree compressed air for another 20 seconds. After the root apexes were sealed with cyanoacrylate (Cyano Anaeróbicos, Buenos Aires, Argentina), the teeth were totally covered with two layers of nail varnish except for a 1-mm peripheral margin on the occlusal P&F. The teeth were then immersed in 1% buffered methylene blue dye solution prepared in artificial saliva (Salivar; Farpag Ltda, Buenos Aires, Argentina) and stored at 37 °C. After 48 hours, the teeth were removed from the dye solution, rinsed in tap water and dried. The crowns were separated from the roots at the cementoenamel junction using a diamond disc. The crowns were then embedded in methyl methacrylate resin and sectioned longitudinally on a buccolingual plane using a diamond wavering blade under watercooling (Isomet, Buehler Ltd, Lake Bluff, IL, USA). A longitudinal section 1 mm thick was obtained at the location of the central fossa.
Evaluation of the sections
After drying with oil-free compressed air, the sections were examined under x14 magnification with stereomicroscope (Carl Zeiss, Oberkochen, Germany) and photographed with a digital Canon Powershot A510 camera (Canon, Tokyo, Japan). Only the side that represented the central fossa was analyzed. The images were downloaded on a computer to enable the length of the P&F to be measured and graded according to the following grading system: C1: P&F extended to half of the enamel thickness; C2: P&F extended beyond half of the enamel thickness without reaching the dentineenamel junction; C3: P&F extended to the enameldentine junction. When a section had more than one P&F, the deepest one was recorded. The images were examined by two independent observers. In case of disagreement, the samples were reexamined jointly until a consensus was reached. The relative frequencies within each group (expressed in percentages) and their corresponding confidence intervals (95%) were then calculated.
RESULTS
All sections revealed the presence of at least one P&F. Descriptive data for P&F classification and frequency along with their confidence intervals are reported in Tables 1 and 2. For pits, 35 and 9 samples corresponded to C1 and C2, respectively (Fig. 1 A and B). Six were considered to be in the C3 category (Fig.1 C). In four of them, incipient traces of dentine dye penetration were observed (Fig.1 D). For fissures, 15 samples corresponded to C1, 18 to C2 (Fig.1 E) and 17 to C3. All samples rated C3 revealed traces of dentine dye penetration (Fig.1 F and G). Based on the 50 studied P&F sections, the enamel-dentine junction was affected in 12% of pits and 34% of fissure samples. Therefore, the null hypothesis was rejected.
DISCUSSION
The present study aimed to mimic some of the in vivo conditions in which the enamel is permanently challenged by the oral environment. For this study, the samples were exposed to a methylene blue dye solution prepared in artificial saliva which contained potassium, magnesium and calcium electrolytes, though without bacteria, unlike intra-oral saliva. In order to estimate the frequency and depth of P&F, the methylene blue dye solution was used as the marker. In previous ex vivo studies 1-4 , dye leakage was used as a predictor of bacterial penetration. However, its correlation with clinical implications should be interpreted with caution and still requires further research. The results of this ex vivo study showed that the enamel-dentine junction was affected in less than 50% (46/50) of the study samples. From an epidemiologic point of view, this proportion of P&F that reached the enamel-dentine junction is a suitable indicator of the risk of developing dental caries. Thus, we may speculate that in cases of C3 categories, penetration of bacteria into dentinal tubules may occur, leading to further access to the pulpal chamber 7 , 8 .
Due to the presence of fermentable carbohydrates and bacterial by-products in the dentinal tubules, inflammatory pulp reactions causing pain are frequently reported by patients 7 . Björndal and Mjör 8 emphasized that the internal morphology of the interlobar grooves, as well as the potential depth of P&F, are important factors to consider during clinical examination. However, Ricketts et al 11 reported on the difficulty to detect enamel alterations or incipient caries formation in the deepest part of the P&F because the morphology of enamel does not always follow a normal pattern 12 , 13 .
Our results are in agreement with previous reports 10 , 14 which suggest that every effort should be made to ensure early detection of decalcified areas 14-16 , which is a consequence of cariogenic onset due to active acid-producing plaque at the base of P&F. Careful examination of the occlusal anatomy will enable the clinician to select the most effective treatment for P&F 4 , 17-20 and make the decision to restore with a filling or apply a P&F sealant. Unprotected (unsealed) P&F can act as an open gate for the entry of bacteria and other irritants from the oral environment. However, since this study was performed under ex vivo conditions, its clinical relevance needs to be interpreted with caution.
CONCLUSION
Within the limitations of the present study, we conclude that in human mandibular third molars, the P&F frequently reach the dentin-enamel junction. When this is the case, there is a potential open gate for bacteria and other irritants from the oral environment to penetrate the dentinal tubules, thereby enabling bacteria and bacterial toxins to reach the dental pulp. This may explain the clinical symptoms reported by patients in otherwise healthy teeth.