1. Introduction
Preventing disease or interrupting its progression at the earliest possible stage while preserving the maximum amount of healthy, viable tissue is a fundamental principle of minimally invasive dentistry [
1]. In this context, vital pulp therapy (VPT) prioritizes the preservation of dental pulp and its physiological functions, including proprioception, innervation, and vascularization, which are lost following root canal therapy (RCT) [
2,
3]. Choosing between VPT and RCT depends on the careful consideration of the predicted success rate in an individual case. Various factors might influence the choice of VPT modality and its overall outcome, such as adherence to aseptic protocols, the preoperative inflammatory state of the pulp, and selecting an appropriate pulp capping material [
4]. However, statistically sound evidence of the effect of patient-related factors remains sparse, with a significant gap in our understanding of how they interact to influence treatment outcomes.
VPT is a less technically demanding procedure, requiring less time and cost than RCT, making it a more accessible and efficient treatment alternative in appropriate cases [
5,
6]. Furthermore, RCT-treated teeth exhibit a higher risk of long-term complications, such as increased susceptibility to fractures, which can compromise the longevity of the tooth [
7,
8]. By preserving pulpal vitality, VPT enhances the tooth’s natural defense mechanisms and supports continued dentinogenesis, contributing to better long-term prognoses [
9]. A systematic review in 2024 [
10] indicated that the success rate of VPT in decayed mature teeth was >90%.
Despite the benefits of VPT, accurately diagnosing pulpal inflammation remains challenging. The traditional classification system of the American Association of Endodontists (AAE) relies on a binary distinction between reversible and irreversible pulpitis, often leading to misclassification and overtreatment [
11,
12]. Wolters et al. (2017) proposed a diagnostic system based on symptoms, categorizing pulpitis into four stages: initial, mild, moderate, and severe, with pulpal exposure and pulpotomy indicated only in select categories [
13]. However, only a few studies currently examine its efficacy in predicting the outcomes of VPT procedures [
14].
Another critical factor influencing the success of VPT is the selection of pulp capping materials, which must provide biocompatibility, antimicrobial activity, and a stable seal against bacterial microleakage [
4,
15]. Mineral trioxide aggregate (MTA) has long been considered the gold standard owing to its excellent sealing ability, high biocompatibility, and capacity for inducing hard tissue formation. However, MTA has notable drawbacks, including tooth discoloration (up to 33% of cases), long setting time (approximately 2–3 h), and difficult handling properties [
16].
In contrast, Biodentine (Septodont Ltd., Saint Maur des Faussés, France) offers improved handling and discoloration resistance, with a reported setting time of 10–12 min [
17]. The success rate with Biodentine in VPT (83–96%) is comparable to MTA [
15,
18,
19]. Additionally, it can be used as a temporary material to fill the cavity, easing its clinical applications. However, its actual setting time is often significantly longer than the manufacturer’s claim, raising concerns about its sealing ability shortly after application [
14].
This pilot study aims to estimate the sample size required for a multivariate regression model assessing how capping material, patient age, sex, and pulp diagnosis influence the success rate of VPT. The findings will also guide the design of more extensive, statistically powered studies to definitively assess the impact of these variables on VPT success.
2. Materials and Methods
2.1. Study Design
A prospective 4-year cohort study was designed to evaluate various VPTs for carious pulp exposure in mature permanent teeth following caries excavation, with the current analysis focusing on the preliminary 6-month outcomes. Ethical approval was obtained from the Medical Research Council (No. 154/13), and all participants provided written informed consent after receiving a full explanation of the study’s purpose and methodology. The trial registration number was NCT06844448.
2.2. Subjects
All adult patients (over 18 years old) who attended the Department of Restorative Dentistry and Endodontics at Semmelweis University (Budapest, Hungary) during the study period (June 2024–January 2025) were examined for inclusion. A patient was included if they had a tooth meeting the following criteria: permanent single- or multiple-rooted teeth with nontraumatic pulp exposure after non-selective caries removal, giving a positive response for dental cold spray (EndoFrost, Coltene-Whaledent, Langenau, Germany), with the absence of any periapical radiolucency on radiographs (periapical index score ≤ 2), restorable, on which absolute isolation can be applied, with a probing pocket depth below 3.5 mm (BPE < 3), and for which mobility is within normal limits (mobility I).
Teeth with a negative response to sensibility testing and pain on percussion, sinus tract, swelling, non-restorable crown, or immature roots and primary teeth were excluded. The signs of severe pulp calcification, internal/external resorption, or pulp exposure due to trauma also formed the basis for exclusion.
Those considered potentially suitable for the study were informed about the procedures’ associated risks, benefits, and alternative treatment options, and subsequently, informed consent was obtained. The consent procedure further explained that the patient would only be enrolled in the study if the pulp was exposed after non-selective caries removal.
2.3. Interventions
After the administration of the local anesthesia (Lidocaine–adrenaline 20 mg/0.01 mg/mL, Egis, Budapest, Hungary) and a mouth rinse with 0.2% chlorhexidine, the tooth was isolated with a rubber dam and liquid dam (Opaldam, Ultradent, South Jordan, UT, USA). The crown was disinfected with a cotton pellet soaked in 5% NaOCl, and under an operative microscope (Flexion, CJ-Optik GmbH & Co. KG, Asslar-Werdorf, Germany), caries was non-selectively removed. The affected enamel was removed using a high-speed diamond bur (558, Jota, Rüthi, Switzerland) with copious irrigation. Subsequently, the gross caries was removed with a steel bur in a slow handpiece from the periphery. The absence of carious dentine on the non-pulpal walls was confirmed using a caries indicator (Sable Seek, Ultradent, South Jordan, UT, USA). The cavity was flushed with NaOCl to decrease the bacterial load as the pulp chamber approached. During bulk caries removal, the stainless-steel bur becomes contaminated by bacteria. Under microscopic magnification, the pulp is generally well visualized, enabling careful cleaning around it while preserving a very thin layer of dentin above. To minimize the risk of pulp contamination, this final layer was removed using a sterile high-speed diamond round bur (801L.FG.012—Medium, Jota, Rüthi, Switzerland). In cases of accidental pulp exposure during bulk removal, the procedure was continued as a partial pulpotomy, given that the exposure occurred in a contaminated environment.
After pulp exposure, different treatment types were used depending on the pulpal status.
Direct pulp capping was applied in the case of no symptoms (i.e., normal pulp) or symptoms of reversible pulpitis when there were no dentine chips in the pulp wound, no signs of necrosis in the pulp, and controllable bleeding within 5 min and the pulp was surrounded with healthy dentine.
After an inspection of the surgical field, the cavity was irrigated with 5 mL of 2.5% NaOCl. A sterile cotton pellet soaked in 2.5% NaOCl was pressed against the exposed pulp to achieve hemostasis. The cotton pellet was removed after every minute to monitor the bleeding time for up to 5 min. After the bleeding stopped, sterile saline was used to flush NaOCl out of the cavity. Biodentine or MTA+ (Cerkamed, Stalowa Wola, Poland) was mixed according to the manufacturer’s instructions and applied with gentle pressure directly onto the pulpal wound. A random allocation was used using envelope randomization. For Biodentine, the entire cavity was filled with the material. MTA+ was applied in a 2 mm layer on the pulpal wound. Once the MTA+ had initially set, it was covered with glass ionomer cement (Ketac Molar; 3M ESPE, Seefeld, Germany) to seal the tooth temporarily.
- 2.
Partial pulpotomy
Partial pulpotomy was indicated when further caries excavation was required after pulp exposure or when dentine chips were present in the pulp wound. Additionally, partial pulpotomy was performed in cases where the pulp was diagnosed as normal or reversibly inflamed but bleeding could not be controlled within five minutes after pulp exposure.
After the pulp was opened, a 2–3 mm piece of coronal pulp tissue was removed with a sterile high-speed bur with water cooling. After an inspection of the surgical field, the cavity was irrigated with 5 mL of 2.5% NaOCl, and hemostasis was performed with gentle pressure on a sterile cotton pellet soaked in 2.5% NaOCl for 5 min. After the bleeding stopped, sterile saline was used to flush NaOCl out of the cavity, and MTA or Biodentine was placed the same way as in the case of direct pulp capping.
- 3.
Complete pulpotomy
Complete pulpotomy was used for reversible pulpitis or normal pulp when anemic, necrotic tissue was found in the pulp or when bleeding could not be controlled within 5 min after partial pulpotomy. It was also applied in cases of spontaneous or constant pain/nocturnal pain.
Complete coronal pulp tissue was excised to the orifice with a sterile high-speed bur and water cooling. Following an inspection of the surgical field, the cavity was irrigated with 5 mL of 2.5% NaOCl, and hemorrhaging was controlled with gentle pressure on a sterile cotton pellet soaked in 2.5% NaOCl for 5 min. When the bleeding had stopped, sterile saline was used to flush NaOCl out of the cavity, and MTA or Biodentine was placed the same way.
Patients were recalled after one week for the placement of permanent composite restorations and pain assessment. In the Biodentine group, 2–3 mm of the material was removed, while in the MTA group, the glass ionomer temporary restoration was removed before final restoration.
2.4. Outcomes
Preoperative pulpal and periapical diagnoses were established through clinical examination, including cold sensitivity testing, palpation, percussion, and periapical radiographs. Patients completed a preoperative questionnaire assessing pain characteristics, including a visual analog scale (VAS) score (0–10). Cavity size was categorized as deep caries, extremely deep caries, or no caries based on the periapical radiograph. “No caries” referred to restored teeth without detectable secondary caries. “Deep caries” involved lesions reaching the inner quarter of dentin, approximately 0.5–1.0 mm from the pulp, with a residual layer of hard dentin. “Extremely deep caries” extended through the full dentin thickness, with inevitable pulp exposure [
20].
At the 6-month recall appointment, a clinical examination was conducted, which included periodontal probing, percussion, palpation, and sensibility testing with cold spray. Additionally, a radiograph was taken to evaluate the status of the apical tissues and the continuity of the periodontal ligament space.
Failure was indicated by the presence of a sinus tract, periapical radiolucency, pain following percussion tests, clinical or radiographic signs and symptoms of root resorption or apical periodontitis, radiological widening of periodontal ligament, or radiographic appearance of bone disturbance or loss.
A favorable outcome was indicated by no pain or other subjective symptoms, radiological evidence of dentine bridge formation, the absence of clinical and radiographic signs of internal root resorption, and a lack of periapical radiolucency or widening of the periodontal ligament apically, as well as no pain following vertical or horizontal percussion.
The primary endpoint of this study was to evaluate the success rate of preoperative pulp diagnosis (i.e., severity of clinical symptoms). Two types of success rates were defined based on the cold pulp test, clinical findings, and radiographic evaluation. The multinomial outcome categorized cases into three groups: (1) success, defined as a positive response to the cold test and the absence of any clinical or radiographic signs of pulp or periapical pathology; (2) uncertain, defined as no response to the cold test, without clinical or radiographic signs of necrosis or infection; and (3) failure, defined as no response to the cold test in combination with clinical or radiographic evidence of infection or pulp necrosis.
The bivariate outcome used a stricter definition of success by classifying uncertain cases as failures.
The secondary endpoint assessed the association between material, age, sex, and VPT modalities. All of the data were collected in the REDCap system (Research Electronic Data Capture-Vanderbilt,
https://redcap.onkobank.com/, accessed on 16 March 2025).
2.5. Statistics
The prevalence of baseline characteristics and the success and failure rate were given as counts in the text and tables, except for the age, which was the mean.
The distribution of sex, American Association of Endodontists (AAE) pulp diagnoses (normal pulp, reversible pulpitis, irreversible pulpitis), Wolters’s diagnostic classification (initial, mild, moderate, severe pulpitis), and treatment type between the two material groups was assessed by Pearson’s Chi-square test. The age was evaluated by the independent t-test.
The multinomial outcome (success, uncertain, failure) was analyzed using Pearson’s Chi-square test without covariates (univariate model) because multinomial logistic regression cannot be computed for zero events. The odds ratio of success versus failure (strict success rate outcome) was analyzed using Pearson’s Chi-square test without covariates (univariate model) and by binominal logistic regression, which comprised the covariates such as age, sex, diagnosis, and treatment types. The two types of diagnosis and the choice of treatment type are not independent factors, so these factors were included in separate models. All statistical calculations were performed using SPSS software (Version 29, IBM Ltd., New York, NY, USA).
2.6. Bias Control Strategies
Multiple measures were integrated throughout the study design, data collection, analysis, and interpretation stages to minimize potential sources of bias.
Broad eligibility criteria were applied, and all consecutive patients who met these criteria during the study period were enrolled.
Examination and treatment procedures were standardized. All clinical procedures were performed by an experienced operator, who had completed postgraduate training in VPT. This eliminated inter-operator variability and ensured consistency in clinical decision-making and technique application. The operator underwent calibration training using 30 independent cases unrelated to the study cohort. The training included two evaluation sessions with a 30-day interval, achieving an intra-examiner reliability kappa coefficient of 0.915 for primary outcomes.
Interpretation bias was mitigated by predefining outcome criteria and conducting blind statistical analyses. The statistician performing the data analysis was blinded to the treatment groups, ensuring an objective interpretation of the results.
3. Results
Fifty patients were screened for suspected pulp chamber exposure during carious tissue removal or symptomatic pulpitis management. Of these, 11 were excluded due to a preference for RCT/extraction (
n = 6), refusal to participate (
n = 2), or failure to meet inclusion criteria for extremely deep caries (
n = 3). Consequently, 39 teeth were enrolled, with 12 subsequently excluded intraoperatively. Among these 12 cases, 4 lacked carious pulp exposure after non-selective caries removal and received either indirect pulp capping or direct composite fillings. The remaining eight cases were treated immediately with RCT due to unsuccessful hemostasis within 5 min at the orifice level (
n = 3) or pre-existing necrotized root canals identified upon pulp chamber entry (
n = 5). Ultimately, 27 teeth were included and assessed after 6 months. The baseline variables were not significantly different between the two material groups—Biodentine (
n = 14) and MTA (
n = 13) (
Table 1).
The success rate was 64% in the Biodentine group and 92% in the MTA+ group. However, the number of uncertain cases was as high as 21%. No significant difference was found between the two groups in the odds ratios either in the multinomial or binomial outcome (
Table 2). However, a notable difference was observed in tooth discoloration, as MTA+ caused discoloration in 8 out of 13 cases, whereas Biodentine resulted in no discoloration in any of the 14 cases
In multivariate logistic regression, the OR (0.14) was similar to the univariate analysis (0.15), indicating no significant effect of covariates (diagnosis, sex, and age) on the outcome (
Table 3,
Table 4 and
Table 5). Although the odds of failure were 6.6 times higher in the Biodentine group, the ratio was not significant.
The increasing severity of the pulpitis with either diagnosis method tended to decrease the success rate (lowering the OR). Similarly, the rising invasiveness of the treatment resulted in a lower odds ratio (favoring failure).
Sample Size Calculation
A sample size calculation was performed based on the parameters of the univariate logistic regression model with predictors of Wolters diagnosis. The strict success rate (bivariate) was used as a dependent variable. The OR of severe pulpitis versus initial pulpitis (0.48) was used for the calculation. Using a significant level of 0.05, 80% power, R2: 0.00, an event rate of 0.3, and a two-tailed test, the required sample size was found to be 140 participants. This calculation ensures sufficient power to detect the specified effect with the parameters for long-term results.
4. Discussion
This study revealed no significant difference in the success rates between the MTA+ (92%) and Biodentine (64%) groups. Both materials have demonstrated efficacy in treating various exposed mature permanent teeth. The results with MTA+ align with previous studies using the widely used ProRoot MTA. MTA+, a cost-effective alternative, retains the proven benefits of MTA, which is widely regarded as the gold-standard material for VPT due to its excellent sealing ability, biocompatibility, and long-term clinical success [
15]. Taha et al. reported a 91.8% success rate for complete pulpotomy in mature permanent molars using ProRoot MTA [
21]. Similarly, Uesrichai et al. [
22] achieved a 92% success rate for partial pulpotomy in maxillary and mandibular molars with irreversible pulpitis using the same MTA. A recent meta-analysis [
15] confirmed the non-inferiority of Biodentine (77.9%) compared to MTA (92%). Both bioactive cements exhibit comparable biological properties for VPT, including high biocompatibility, odontogenic potential, low inflammatory response, angiogenic activity, and antibacterial effects [
22]. However, despite these shared biological characteristics, they differ in physical properties such as handling, setting time, and cost, which may influence material selection in clinical practice [
14].
While both materials demonstrated favorable outcomes, MTA tended to have a higher success rate in this study. The overall clinical success rate in the first six months for the Biodentine group was lower than the success rates reported in previous studies [
21,
22]. This discrepancy may be attributed to the strict success criteria applied in this investigation, such as the classification of negative cold response cases as a failure. Biodentine is suitable as a single-visit VPT material with a claimed setting time of 12 min. However, despite adherence to the manufacturer’s instructions, its actual setting time often extends two to three times longer [
23,
24], averaging around 22 min [
14]. This delayed setting might reduce clinical efficiency, as it is impractical to place the definitive restoration in a single appointment, thus diminishing the advantage of Biodentine over MTA, which sets in several hours.
Notably, unlike other MTA materials, Biodentine was not associated with postoperative discoloration [
14,
25]. This aligns with our findings, where no samples in the Biodentine group exhibited gray discoloration, unlike several cases in the MTA+ group.
This study found a high success rate for all three VPT techniques, which aligns with the other studies [
14,
26,
27].
Multivariate analysis showed that complete pulpotomy had a slightly higher success rate than partial pulpotomy; however, this difference was not statistically significant due to the small sample size and the multifactorial nature of the outcome. Similarly, Louzada [
28] and Jakovljevic et al. [
12] found no significant differences among the groups but a tendency of a decreasing success rate with the more invasive treatment. In contrast, Brodén et al. [
29] concluded that no definitive preference between direct pulp capping and partial pulpotomy could be established. However, we applied direct pulp capping in patients with no pulpal symptoms diagnosed as having normal pulp, where the pulp was exposed during deep caries removal, which could result in a higher success rate. In contrast, patients with less severe initial pulpal conditions were treated with partial or complete pulpotomy. The initial diagnosis of the severity of the pulpitis was supported by the presence and extent of deteriorated pulp tissue by microscope. Therefore, it is not surprising that the success rates of VPT techniques were inversely correlated with the invasiveness of the procedure. This observation is consistent with the findings of Zhang et al. [
30]. Careddu et al. found the same trend: mild pulpitis had a 100% success rate, moderate pulpitis 88%, and severe pulpitis 60% at 1 year [
14]. These findings suggest that the Wolters classification could predict treatment failure. Similarly, irreversible pulpitis is classified by the ESE terminology, resulting in a higher OR than reversible pulpitis [
31].
As the depth of caries and the diagnosis are confounded with the treatment modality, a prospective cohort study is more advantageous (and ethically sound) than a randomized control study to determine the coefficient of the predictors.
Defining success criteria, such as the response to sensibility testing, is crucial as it significantly impacts reported treatment outcomes. Cold testing may underestimate pulpotomy success due to potential false-negative responses. [
14,
32]. For example, a study by Careddu and Duncan reported a 90% success rate for partial pulpotomy, including both “successful” and “unresponsive but successful” cases [
14].However, when unresponsive cases were classified as failures, the success rate dropped to 66% [
14]. Similarly, in this study, partial and complete pulpotomy cases frequently did not respond to cold testing even though the tooth was asymptomatic and showed no evidence of apical radiolucency. The European Society of Endodontology (ESE) recommends a positive response to pulp sensibility testing to indicate success. However, it acknowledges that in certain cases, particularly in complete pulpotomy, a positive response may not be expected [
20]; therefore, complete pulpotomy cases were considered successful despite the negative reaction. Due to the short observation period, the strict criteria for the classification of these unresponsive cases as failures were used in this study.
Moreover, electric pulp testing (EPT) is often used as an alternative, showing more positive responses, particularly in partial pulpotomy cases, where the success rate for EPT was 95.45% compared to 45.45% for cold testing [
32]. However, this approach reveals the inconsistency in the literature regarding the classification of unresponsive cases, with some studies considering them successful despite the lack of response [
14]. This underscores the need for further investigation and standardized criteria to accurately assess the success of VPT procedures, especially considering the limitations of cold testing in these cases.
This study’s strength lies in its rigorous methodology, including strict inclusion criteria, decision-making, a standardized follow-up protocol, and systematic evaluation criteria to ensure objective and reliable outcome assessments. Unlike previous studies focusing on single diagnostic systems, multiple classification frameworks were analyzed, allowing for a broader comparison of clinical relevance and prognostic accuracy. Additionally, by including all stages of pulpal inflammation, this study provides a comprehensive understanding of pulp pathology and treatment outcomes, enhancing its applicability in VPT. Moreover, one experienced operator performed all the treatments, which excludes differences due to the operator. Another strength of this study is that it includes only permanent teeth of adults without limitations regarding age.
This study had limitations, including a small sample size, limited follow-up, and the use of cold spray as the sole method for sensibility testing. Full pulpotomy rarely responds to cold tests; the success criterion of a “positive response to cold” cannot be applied to this treatment modality. Case selection posed another challenge, as full pulpotomy was only performed in cases of irreversible pulpitis. In contrast, for teeth with normal pulp, exposed pulp, or reversible pulpitis, direct pulp capping, partial pulpotomy, and full pulpotomy were all performed.
Additionally, the lack of randomization in the treatment modality level could have introduced bias; however, proper randomization was not feasible due to the nature of the treatment.
There is a need for more uniform and comprehensive approaches to assessing the efficacy of VPT techniques and to standardize success criteria for a more accurate comparison and interpretation of results. Combining cold and electric pulp testing would enhance the assessment of pulp vitality and treatment outcomes.