Selective stepwise caries removal in primary teeth: a microbiological assessment on surviving microbiota

Partial caries removal performed by the stepwise treatment has a high rate of clinical success and promotes a reduction of microorganisms in carious dentin. However, the adaptive behavior of these cloistered bacteria is not entirely clear. Aim: This study aimed to evaluate the carious dentin and quantify the microorganisms Streptococcus mutans and Lactobacillus acidophilus at the first intervention and after 90 days, assessing the acidogenicity and aciduricity of these bacteria isolated from the lesions. Methods: Twenty patients presenting deep caries lesion in primary molars eligible to receive the stepwise treatment were selected, dentin samples were collected in two different moments: in the first intervention, just after partial caries removal; and in the second intervention, during the reopening of the cavity (90 days after the temporary sealing of the lesion). The samples were processed for microbiological analyses via culture, identification and quantification. The bacteria isolates were subjected to phenotypic tests of acidogenicity and aciduricity. Dentin consistency and color was also recorded by a calibrated examiner. Data were statistically analyzed. Results: There was a reduction in the number of viable microorganisms while dentin rehardening and browning was noted (p<.05), but no change occurred in the acidogenicity and aciduricity properties of Streptococcus mutans and Lactobacillus acidophilus over time. Conclusion: Thus, the stepwise treatment promoted clinical changes as darkening and hardening of carious dentine and promoted a reduction in the number of viable microorganisms, but no influence was found on the phenotypic characteristics of acidogenicity and aciduricity of the species analyzed after 90 days.


Introduction
It is no longer a novelty that the term "minimally invasive" is almost a synonym of contemporary dentistry. That is why the mechanical removal of carious tissue becomes the "least as possible". The complete removal of the demineralized dentin to reach sound dentine is considered an unwise practice . Hence, partial removal of decayed tissue through selective excavation might minimize pulp damage, giving it the chance of recovery. Nonselective removal of carious tissue, which consists of removing all softened carious dentin, presents a greater risk of pulp exposure, and it might be considered overtreatment (Casagrande et al., 2017). Pulp exposure may initiate a cascade of re-interventions and numerous complications, resulting in the caries process resolution delay and frequently demanding supplementary longstanding endodontic intervention (Pratiwi et al., 2017).
The principle of the selective stepwise treatment is the certification of the cleanliness of the surrounding walls of the cavity and removal of the real softened dentin from the pulp wall, reducing the risk of pulp exposure and stimulating tissue rehardening (Bjørndal et al., 1997). The later reopening will confirm the dentin mineralization and hardness increase at the bottom of the cavity, allowing treatment conclusion by definitive restoration. Selective stepwise treatment is a widely cited procedure with extensive supporting scientific evidence Bitello-Firmino et al., 2018;Maltz et al., 2011). Stepwise excavation is also considered a suitable model to determine the persistent bacteria and their metabolism in clinically excavated lesions (Bjørndal et al., 1997), even though it is considered unnecessary, in some cases, and outdated in view of the selective technique (Elhennawy et al., 2018) The sealing of carious dentine resulted in the same bacteria count after 3 months compared to the complete removal of the affected dentin (Bitello-Firmino et al. 2018). A decrease in viable anaerobic and aerobic bacteria was observed when this tissue was reassessed after 6 months (Maltz et al., 2012). However, 25.6% of teeth reveal bacterial growth even after complete direct excavation; thus, assumptions that this technique eliminates all bacteria present might not support it (Bitello-Firmino et al. 2018). So, the concept of infected or affected dentin appears to be old-fashioned and restricted for a histopathologic approach; therefore, the visual description based on hardness and color is a lot more tangible, clinically (Barros et al., 2020).
Defining the dentin condition, which could differ depending on distinct areas of the cavity, is a difficult task. For practical reasons, checking if the tissue is soft at any spot is of great guidance in the decision-making process "need to be removed or not" (Innes et al., 2016).
The clinical change of carious dentin appearance after the period of cavity sealing is widely noted (Lula et al. 2011;Duque et al., 2009). Although the persistent presence of microorganisms after carious dentin closure has been reported (Maltz et al., 2012;Duque et al., 2009;Lula et al., 2009), the adaptive behavior of these cloistered bacteria is not entirely clear. The sealed cariogenic bacteria, deprived of dietary carbohydrate, are provided with serum proteins from pulp fluids (Ganas et al., 2019;Knutsson et al., 1994). This nutrition seems to be enough to keep producing metabolites and organic acid somehow (Ganas et al., 2019), and that is one of the supporting evidences on the recent bioactive restorative material development, mainly those with antimicrobial additives (Tüzüner et al., 2019). However, the relative simplicity and homogeneity of the nutrient supply significantly affected the surviving microbiota, and the microbiome is less complex regarding phenotypic and genotypic characteristics (Paddick et al., 2005). Just one study has observed the phenotypic characteristics of these bacteria after this fasting period (Paddick et al., 2005). It is intriguing how fasting can affect microbiome, some influence on gut microbiome has already be investigated (He et al., 2019). Therefore, the present study investigated dentin color and hardness alteration, microorganisms' survival, and acidogenicity (acid production) and aciduricity (acid tolerance) properties of mutans streptococci and Lactobacillus spp. after selective stepwise removal of carious dentin. The null hypothesis is there is no alternation in microorganisms aciduric and acidogenic properties after 90 days of the temporary sealing of the caries lesion.

Sample Selection
After screening 145 patients referred to the Child Dental Care Service at Araraquara School of Dentistry (São Paulo State University, Brazil), twenty patients requiring indirect pulp treatment for primary posterior teeth, diagnosed after clinical and radiographic examination, were included in the present study. The Institutional Ethical Committee approved the study protocol (CAEE: 13823613.5.0000.5416) and informed consent was obtained from all volunteer legal guardianship. All patients (selected and non-selected) were treated at Araraquara School of Dentistry following the standard treatment adopted by the institution, based on the scientific evidence available at the time.
Participants were selected considering the following inclusion and exclusion criteria. The sample calculation was Inclusion Criteria: Patients with at least one deep caries lesion (active) limited to the occlusal surface and with a risk of pulp exposure during the excavation in the first appointment. Patients aged between 4 to 9 years, without systemic impairment, without report of antibiotic use on the last 3 months previous to screening, without tooth spontaneous pain, tenderness to pressure, mobility, pus formation, and swelling. Radiographically, teeth should have shown radiolucency in the internal half of dentin and absence of internal and/or external dentin resorption and radiolucent area in bi/trifurcation. Volunteers who showed dental pulp exposure at any step of the study were excluded and those with suspected noncompliance with treatment or where it was not possible to recover the analyzed microorganisms.

Clinical Procedures
Patients who matched inclusion and exclusion criteria underwent indirect pulp treatment with stepwise excavation (not only for the purpose of research). The clinical procedures included local anesthesia (Mepiadre 2% 1:100.000 -Nova DFL, Rio de Janeiro, Brazil), operative isolation with a rubber dam (Sanctuary -Kdent/Quimidrol, Joinville, Brazil), removal of affected enamel (if necessary) to allow access to the dentinal lesion, complete removal of carious dentin on lateral surfaces, and partial removal (only softened tissue) on pulp surface. Curettage was interrupted when tissue was slightly resistant to instrument (Massara et al., 2002). The cavity was washed (0.9% NaCl) and dry (absorbent wipe). Dentin chips were collected from the pulp surface and immediately stored in a microtube filled with phosphate buffer solution (PBS-1 mL, pH 6.8-7.2), previously weighted. The cavity was filled with restorative glass ionomer (Ketac Molar Easymix -3M ESPE, Nova Veneza, Sumaré, SP, Brazil) and protected with varnish.
After 3 months, restorative material was removed. The pulp surface was assessed, and another dentin fragment was removed and stored as described previously. Caries lesion was finally sealed with glass ionomer (Vitrebond -3M ESPE, Nova Veneza, Sumaré, SP, Brazil) followed by resin composite (Z350 -3M ESPE, Nova Veneza, Sumaré, SP, Brazil). All interventions were performed at Araraquara School of Dentistry (São Paulo State University, Brazil).

Dentin Consistency and Color
All procedures described were carried out by a single investigator who was the same calibrated examiner for dentin texture/hardness and coloration assessment (Kappa -0.84 and 0.80, respectively). Dentin appearance was evaluated twice, at baseline and 3 months after the baseline. The aspects analyzed were based on Maltz et al., 2002. The dentin color was classified as yellow and light brown. The dentin consistency was classified as soft (little resistance to removal), leathery (greater resistance to removal, being removed with some firmness), and hard (a consistency similar to healthy dentin).

Microbiological analysis
The microtubes containing 1 ml of PBS were weighed before and after placing the dentin chips inside to obtain the weight (Scale AY220, Shimadzu, Barueri, Brazil) of moist dentin in mg. Samples were sonicated for 5 seconds under 20W For aciduricity analysis, all strains were adjusted to contain approximately 10 7 CFU/mL through optical density (OD -Streptococcus mutans = 0.15 to 0.20 and Lactobacillus acidophilus = 0.50 to 0.60) (Biophotometer, Eppendorf, Hamburg, Germany). All microbial suspensions were centrifuged and washed with 0.1 M glycine buffer (pH ~7.0). Next, the pellets were resuspended in 0.1 M glycine buffer at pH 3.0 and 5.0. Serial dilution was carried out immediately after resuspension (T0), and 30 (T1) and 60 (T2) min after incubation at 37 o C. Samples were plated on BHI Agar (Neogen, Indaiatuba, Brazil) Petri dishes and incubated at 37 o C for 48h. Bacterial viability was counted and expressed in CFU/mL. The samples that grew at pH 7.0 were considered as controls.
For acidogenicity evaluation, microorganisms were adjusted as described previously for aciduricity analysis, the samples were resuspended in PBS (pH 7.0), and then glucose at 55.6 mM was added. The pH of the solution was measured four times (baseline, and after 60, 120, 180 min) using a previously calibrated electrode (pH meter 3510 -Jenway, Staffordshire, UK).

Statistical analysis
All data were analyzed regarding assumptions of normality (Shapiro-Wilks test) and homogeneity of variables (Levene test). The CFU/mg values were later transformed in Log10. Wilcoxon nonparametric test was used to identify significant differences in dentin consistency, color, and CFU/mg of moist dentin. For acidogenicity and aciduricity statistical analyses, the two-way ANOVA with repeated measures was applied. Statistical analysis was performed using the SPSS software (SPSS Inc, Chicago, IL), considering a significance level of 5%.

Results
Among the 20 patients included in the study, 9 were males and 11 were females. The age ranged from 4 to 9 years old with the mean age of 5.7 ± 1.4. The teeth distribution consisted of 13 primary second molars (6 upper and 7 lower) and 7 primary first molars (3 upper and 4 lower). The average time between the first and the second intervention was 93 ± 7.6 days.
No pulp alteration was identified clinically or radiographically, nor pulp exposure or restoration loss.
Alteration regarding color and consistency of the remaining dentin in both moments are shown in Table 1. The selective stepwise removal of carious dentin promoted significant changes in dentin consistency and color after 3 months, reaching a darker and harder aspect. A significant decrease in all microorganisms cultivated between the first intervention and 3 months later was observed (Graph 1). Regarding the phenotypic tests of Streptococcus mutans and Lactobacillus acidophilus, the results of acidogenicity are found in Graph 2 and 3. A similar neutral pH was obtained for all species in the baseline and a significant and consistent reduction was observed later on. Both species showed a significant drop after 60 minutes of measurement. The samples collected from patients presented a lower pH in comparison with ATCC strain in all times set. For Lactobacillus acidophilus, no difference was observed between the samples collected from patients in all moments measured. The patients' samples collected in the first intervention showed a statistical lower pH compared to the sample collected 3-months later, only in the measurement at 60 min, for Streptococcus mutans. This found was not observed afterwards.
Graph 2. The relationship between the pH measured for Streptococcus mutans (mean and standard deviation) and time.
Graph 3. The relationship between the pH measured for Lactobacillus acidophilus (mean and standard deviation) and time.
* Indicates difference between the period of analysis. One-way repeated measures -ANOVA test (p<.05). Source: Authors.
The results obtained on aciduricity for Streptococcus mutans and Lactobacillus acidophilus can be seen in Table 2 and 3, respectively. A greater reduction in microbial viability was observed when ATCC strain was tested in both pH (5 and 3), at 30 and 60 minutes in comparison with the samples collected from patients. There was no statistical difference between the viability of the species collected from patients in the first intervention with those collected 3-months later. Streptococcus mutans and Lactobacillus acidophilus viability decreased proportionally as a function of the incubation period. However, for Lactobacillus acidophilus collected from patients, this significant decrease can be observed only after the 60-minute incubation period. For Streptococcus mutans, a significant reduction can be observed even after 30 minutes, but only at pH 3.

Discussion
This study investigated the possible basis for observations that a cloistering and fasting period is able to change phenotypic characteristics of a cariogenic biofilm. The number of viable bacteria was also counted before and after selective stepwise treatment. Dentin consistency and color was assessed as a method of confirming the effectiveness of the treatment protocol, even in the face of the consistent evidence available to support it (Barros et al. 2020).

Streptococcus mutans and
Lactobacillus acidophilus showed lower, but not significant, capacity of acid production after 3 months of stepwise treatment.
Whereas aciduricity changes was found to be not significant neither, then the hypothesis that selective stepwise removal of carious dentin could impact phenotypic aspects of microorganisms is not accepted.
It is important to note that selective removal of carious dentin may be accomplished in one or two-step protocol, the reason for taking this second (also known as stepwise excavation) in the present study was to allow the phenotypic investigation proposed. The stepwise technique removes caries in stages over two visits some months apart (3 months in the case), permitting the dental pulp time to lay down reparative dentine. There is no sufficient evidence to support one protocol instead of the other (Ricketts et al., 2013), but the cost-effectiveness favors the one-visit treatment . It is also important to highlight that the study was carried out in deciduous teeth which have a higher capacity for cell proliferation, response, and regeneration (Kaukua et al, 2015), extrapolation to permanent dentition should be carefully investigated beforehand.
The change in dentin colours and consistency observed in this study is attributed to biochemical reactions, like the Maillard reaction, and the reaction of small aldehydes derived from bacteria with proteins forming polymers with a more brownish color (Kleter et al., 1998). A negative correlation between carious dentin consistency and S. mutans has already been suggested (Lula et al., 2011). Once the reduction of viable total bacteria was found in the present study after 3 months, the process of hardening and browning of dentin is expected to be reduced later on.
Several studies have also demonstrated a significant reduction in the quantity and diversity of the collected dentin microbiota after minimally invasive treatment (Duque et al., 2009;Lula et al., 2009;Orhan et al., 2008;Wambier et al., 2007), corroborating the results displayed in Graph 1. Although none of the treatment methods completely eliminates the viable microorganisms, results suggest a higher reduction after reopening in the two-visit treatment (Lula et al., 2009;Ganas et al., 2019;Orhan et al., 2008;Wambier et al., 2007) and no difference between selective and complete removal of carious dentin (Bitello-Firmino, 2018) after 3-months follow-up. The bacterial reduction found is associated to nutritional restriction and antibacterial properties of lining materials, both dependent on adequate marginal sealing of the cavity.
A dysbiotic imbalance within the plaque microbiome is known as the most relevant factor in the dental caries etiology, that leads to an acidic environment as a result of the metabolism of carbohydrates. The ability to survive within this acidic environment should be directly link to the ability to produce acid (Bana et al., 2019). Therefore, a coherence between the acid tolerance and acid production was previously speculated, and later confirmed by the present study.
The permanence of viable microorganisms in the dentin after sealing the cavity, even a smaller amount, is still considered a dilemma. There was no evidence available to define the phenotypic characteristics of cariogenicity of these remaining bacteria. To the best of our knowledge, there is only one study that analyzed the capacity of the remaining microbiota for the production of glycosidic enzymes (Paddick et al., 2005). These enzymes are able to obtain sugars from glycoproteins to keep bacterial metabolism active even with nutritional restriction, conserving their aciduric and acidogenic properties due to the metabolization of sugars from other sources. Thus, the data obtained in this study provide additional evidence to Paddick and collaborators' investigation. The aciduric and acidogenic properties of Streptococcus mutans and Lactobacillus acidophilus collected before and after the stepwise excavation did not differ statistically from each other, difference was observed only when compared to their standard strains.
Considering that the analyzed bacterial species are reduced after stepwise excavation, but still present, and the phenotypic characteristics of the microorganisms collected at the beginning of the treatment are maintained, in the presence of an eventual marginal microleakage, a dentin recolonization might not be necessary for caries process turnover, only nutrients availability is needed. These results highlight the importance of an appropriate marginal sealing for the success of selective removal of caries.

Conclusion
Although a reduction of viable microorganisms was observed after 3-months of selective stepwise excavation, those that still remain have the ability to ferment and survive within an acid medium similar to the microorganisms collected in the initial intervention.