Five-year clinical outcomes of root canal treatment using a calcium silicate–based sealer in teeth with apical periodontitis: a retrospective cohort study

Abstract

Objective

Hydraulic calcium silicate–based sealers (CSBSs) are increasingly used in endodontics, but evidence on their long-term clinical effectiveness remains limited. This retrospective study evaluated the 5-year outcomes of primary root canal treatments and retreatments performed with a CSBS in teeth with symptomatic apical periodontitis.

Materials and methods

The sample consisted of 637 teeth with symptomatic apical periodontitis treated by a single experienced operator using a premixed CSBS (CeraSeal; Meta Biomed Co., Cheongju, South Korea) with the cold hydraulic condensation technique (CHC). Data were retrospectively collected from clinical charts and radiographic archives. Preoperative variables included sex, systemic health, tooth type, arch location, lesion size, pulpal diagnosis, sinus tract, and preoperative small perforation. Intraoperative factors comprised intracanal medication, apical diameter, obturation length, and sealer extrusion. Postoperative information included the type of coronal restoration. Follow-up evaluations were based on available radiographs and clinical notes at 1, 2, 3, 4, and 5 years. Outcomes were classified according to strict criteria (absence of symptoms and complete radiographic resolution) and loose criteria (absence of symptoms with complete or partial healing). Prognostic factors were investigated using bivariate associations and multivariate logistic regression models.

Results

Strict success rates increased progressively (80.2% at 1 year, 85.7% at 2 years, 86.8% at 3 years, 88.2% at 4 years, and 87.67% at 5 years). Loose success remained consistently high (99.2% at 1 year; 90.7% at 4 years, and 87.67% at 5 years). Higher baseline periapical index (PAI) scores significantly reduced odds of strict success at 2, 3, 4 years. Other factors, including age, sex, tooth type, treatment type, extrusion, and restoration, were not significant.

Conclusions

CSBSs demonstrated favorable long-term outcomes. Preoperative lesion severity was the main prognostic factor, while demographic and procedural variables showed limited influence on long-term success.

Clinical relevance

This study supports the long-term clinical reliability of CSBSs used with CHC in teeth with apical periodontitis. Preoperative lesion severity remains the primary prognostic factor influencing periapical healing over time.

Introduction

Apical periodontitis (AP), a chronic inflammatory disease of periapical tissues resulting from intracanal infection, is a frequent radiographic finding in both untreated and previously treated teeth [1]. Epidemiological data estimate that approximately 52% of the adult population harbors at least one tooth affected by AP, and approximately 39% of these cases involve teeth that have already undergone endodontic treatment [2]. The resolution of AP is closely related to the elimination of intracanal infection and the quality of the final root canal filling [3].

Creating a hermetic seal of the root canal space is a crucial step for the success of non-surgical canal treatment and retreatment [4–6]. Many studies have found that the presence of apical periodontitis significantly decreases the success rate of endodontic treatments, especially in retreatments [7–14]. Hydraulic calcium silicate–based sealers (CSBSs), particularly in their premixed form, have emerged as a promising alternative to traditional obturation materials since they exhibit advantageous biological and physicochemical properties, including biocompatibility, bioactivity, antibacterial potential, chemical stability, low solubility, and favorable flowability [15–19]. Although the evidence supporting the effectiveness of root canal filling materials and techniques in the treatment of apical periodontitis remains limited [20], CSBSs have shown promising clinical outcomes, with success rates comparable to those reported for conventional sealers [21].

CSBSs demonstrate many desirable properties for their use in clinical practice, such as biocompatibility, bioactivity, chemical stability, sealing ability, antibacterial properties, low solubility, flowability, dentinal tubules penetration, no tendency to shrink, calcium ion release, and hydroxyapatite formation [22]. These qualities allow them to be used with the cold hydraulic condensation (CHC) technique, using a well-fitting cone in round canals (single cone technique), or passively adding other cones in oval canals; the technique is time-saving [23], and aims to create a biological seal [24–28].

Precisely, the CHC technique, when used with CSBSs, has demonstrated clinical outcomes comparable to those achieved with thermoplasticized techniques, including warm vertical compaction and carrier-based systems, with no significant differences in long-term healing rates [29, 30]. From a clinical perspective, the technique offers several practical advantages, including reduced technique sensitivity, greater reproducibility, and shorter working time, as it relies on sealer flow rather than operator-dependent compaction forces [31]. In addition, the bioactive properties of CSBSs, including ion release and apatite formation, may compensate for the absence of thermoplasticized gutta-percha and contribute to effective sealing and periapical healing [31].

Among these materials, CeraSeal (Meta Biomed Co., Cheongju, South Korea) is a premixed CSBS composed of calcium silicates, tricalcium aluminate, zirconium dioxide, and proprietary thickeners [22, 32, 33]. Its ability to release calcium ions and induce hydroxyapatite formation may contribute to enhanced sealing and periapical healing [34]. Due to its recent release, knowledge about this material in the literature is scarce, especially in terms of clinical studies [32].

Recent studies have begun to investigate the clinical performance of CSBSs, with several reporting comparable success rates and clinical outcomes to those of conventional resin-based sealers [35–37]. However, the existing literature presents some notable limitations. Most published studies include multicenter studies with heterogeneous operator experience, lack of protocol standardization, or are conducted in academic settings under controlled conditions [38]. Moreover, sample sizes are frequently limited, and very few investigations have focused on treatments performed entirely by a single, experienced clinician using a consistent technique across all cases [16, 38, 39]. This lack of methodological consistency reduces the generalizability of existing findings to real-world scenarios. Furthermore, significant differences in the chemical composition of commercially available CSBSs may lead to variations in clinical performance and biological activity [40].

Importantly, to date, no published study has specifically assessed the long-term clinical success rate and prognostic factors in teeth diagnosed with apical periodontitis treated with CeraSeal.

The purpose of this retrospective study was to assess the clinical success rate of non-surgical root canal treatments and retreatments in teeth affected by apical periodontitis, using the CHC technique with CeraSeal sealer in a private practice setting with follow-up periods of up to 5 years.

Materials and methods

Study design and patient selection

This retrospective monocentric study was reported in accordance with the guidelines for reporting observational studies in endodontics (PROBE) [41].

All procedures performed in this study involving human participants, were in accordance with the ethical standards of the institutional committee of the Azienda ospedaliero-universitaria Senese (Siena, Italy) and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed written consent was obtained from all individual participants included in the study. This retrospective cohort study included patients treated in a private practice (Ancona, Italy) between 2018 and 2023 by a single experienced endodontist with over 30 years of clinical experience (F.C.). All cases involved non-surgical root canal treatment or retreatment of teeth diagnosed with apical periodontitis, performed using the CHC with a CSBS (CeraSeal; Meta Biomed Co., Cheongju, South Korea). The study included adults aged 18–80 years, classified as ASA I–II, who attended annually for oral hygiene and X-ray control [42]. Teeth were included if they were diagnosed with symptomatic apical periodontitis on clinical examination, showed periapical bone alterations on preoperative radiographs, and had undergone non-surgical root canal treatment or retreatment with adequate coronal restorations. Additional criteria were the availability of preoperative periapical radiographs, at least 24 months of clinical and radiographic follow-up, and treatment performed by a single experienced operator using a standardized protocol (CHC technique with CeraSeal bioceramic sealer). Both single-rooted and multi-rooted teeth were included in the study.

Only teeth with complete root formation and no history of periapical surgery were considered. Teeth were excluded if they had cracks extending into canal orifices, perforations caused by burs or ultrasonic tips, severe periodontal disease, or evidence of vertical root fracture (a narrow, deep probing defect and/or a J-shaped lesion in a previously treated tooth). In contrast, teeth with foramina larger than size 50, as well as those with ledges or small intracanal perforations created by misuse of manual or rotary files, were included. Teeth with apical diameters greater than ISO size 50 were also included, as all cases presented complete root formation and were managed with conventional orthograde treatment.

The periapical status was assessed by palpation, percussion, and thermal cold testing. Afterwards, teeth showing deep caries, deep restorations, lack of response to cold testing, or painful response to biting and/or percussion or palpation were evaluated as potential cases of AP and underwent additional periapical x-rays using the long cone paralleling technique with a film holder [43, 44].

Sample size

The sample size was estimated based on the expected treatment success rate reported in a previous retrospective study on a CSBS (90.9%) [34]. Assuming a 95% confidence level, the available sample of 637 teeth was considered adequate to ensure a reliable level of precision in estimating treatment outcomes.

Treatment protocol

All teeth were treated by a single experienced operator (F.C.) in single or multiple visits under local anesthesia (4% articaine with 1:100,000 epinephrine) and rubber dam isolation.

In primary treatments, after access cavity preparation, canal scouting was performed using stainless steel K-files (Dentsply Maillefer, Ballaigues, Switzerland), followed by glide path creation with MTwo 10.04 and 15.05 (VDW GmbH, Munich, Germany). Canal shaping was carried out using the ProTaper Gold rotary system (Dentsply Maillefer), operated with an E-Connect endomotor (Eighteeth, Changzhou, China) at 350 rpm and 4 N·cm torque, using a step-down approach according to canal anatomy.

In retreatments after access, previous root canal filling materials and obstructions were removed using rotary files, solvents, and when necessary, sonic (KaVo SONICflex; KaVo Dental GmbH, Biberach, Germany) or ultrasonic tips (P5 Newtron, Acteon Group, Mérignac, France). Once patency was re-established, canal shaping was performed as described for primary treatments.

During instrumentation, canals were irrigated with 5.25% sodium hypochlorite (Niclor 5; Ogna, Muggiò, Italy) using an open-ended needle (Endoneedle, Vedefar NV, Mechelen, Belgium). A final irrigation protocol included passive ultrasonic activation (Ultra-X; Eighteeth) or sonic activation (Eddy; VDW GmbH) with 3 mL 5.25% sodium hypochlorite, 3 mL 17% EDTA (Tubuliclean, Ogna), and 3 mL 5.25% sodium hypochlorite.

Canals were then rinsed with distilled or sterile water to avoid any potential interactions between the final irrigant and the bioceramic sealer [45–47].

Working length was determined using an electronic apex locator (E-PEX, Eighteeth) and confirmed radiographically. In cases with an apical diameter exceeding ISO size 50, final apical enlargement was performed manually using K-files (Dentsply Maillefer).

Canals were dried with sterile paper points. Teeth were treated in single or multiple visits depending on canal conditions. Non-dryable canals were medicated with calcium hydroxide (Calcicur; VOCO GmbH, Germany) for 2–4 weeks, whereas dryable and asymptomatic cases were completed in a single visit.

Obturation technique

On obturation visit, a master cone that fits snugly at the working length or at 0.5 from the working length was selected; canals were rinsed with distilled or sterile water to avoid any potential interactions between the final irrigant and the bioceramic sealer [46, 47].

CSBS was delivered directly into the canal using a single-use tip to fill the coronal half. The gutta-percha cone was then slowly and passively inserted to working length and subsequently removed to verify the presence of sealer at the tip. If the tip appeared uncoated, additional CSBS was either delivered into the canal or applied directly to the cone tip, which was then reinserted slowly. If the cone tip was already adequately coated with CSBS, it was reinserted without adding further sealer. In case of oval canals, additional gutta-percha cones were passively added to the master cone.

Gutta-percha cones were seared off at orifice level using a heat source (EQ-V Pack, Metabiomed, Korea; Fast-Pack Eighteeth, China), and then vertically packed with a calibrated plugger. After the obturation, the excess of sealer was removed from the chamber using ultrasonic tips with water, then the orifices were sealed with the permanent composite restoration or with a layer of flow, after etching and bonding, in case of temporary restoration with IRM (Dentsply, York, PA). At the end of the procedure a periapical radiograph was obtained to assess the quality of the root canal fillings. All procedures were performed under magnification 4.8× loupes with light source or surgical microscopes. The sealer used in all cases was CeraSeal (Meta Biomed Co., Cheongju, South Korea). Table 1 summaries the chemical components of CeraSeal. The coronal restoration was completed either with composite resin or a full-coverage crown, depending on clinical indication.

Table 1.

Chemical composition of Ceraseal

Component	Function
Zirconium oxide	Radiopacifier
Thickening agents Tricalcium aluminate Tricalcium silicate Dicalcium silicate	Handling properties and viscosity Setting reaction modulation Primary hydraulic phase, calcium ion release Long-term strength and stability

Exact quantitative composition (weight percentages) is not disclosed by the manufacturer; therefore, only qualitative composition is reported

Follow-up and Outcome Evaluation

The clinical data were collected from the chart records and radiographic database. The preoperative, intraoperative, and postoperative variables were collected to identify any prognostic factors. Follow-up evaluations at 1, 2, 3, 4 and 5 years post-treatment were based on the radiographs and clinical records available in the database. The primary outcome of the study was treatment success, defined as the absence of clinical symptoms and radiographic healing, and evaluated according to both strict (complete healing) and loose (complete or incomplete healing) criteria. Preoperative, intraoperative, and postoperative data were gathered. The preoperative information included sex, controlled systemic disease, treatment type, tooth type, tooth location, lesion size, pulpal diagnosis, sinus tract, and preoperative small perforation in previously treated cases. Intraoperative data recorded included intracanal medication, apex size, and sealer extrusion. Postoperative data comprised the type of coronal restoration. Clinical assessment included the presence or absence of pain during the function, swelling, sensitivity to palpation and percussion, presence of sinus tract.

Standardized periapical radiographs were obtained and independently assessed by two calibrated examiners for the presence, absence, and change (increase/decrease) in size of any periapical radiolucency and for assigning the PAI score at recalls as follows:

• PAI 1: Normal periapical structure.

• PAI 2: Bone structural changes indicative but not pathognomonic for apical periodontitis.

• PAI 3: Bone structural changes with some mineral loss characteristic for apical periodontitis.

• PAI 4: Well-defined apical radiolucency.

• PAI 5: Radiolucency with radiating expansion of bone structural changes.

Healing was determined as a decreased PAI score and lack of symptoms.

Treatment outcomes were categorized as healed, diseased, or healing based on clinical findings and radiographic appearance. For statistical analysis, outcomes were dichotomized into “success” (healed or healing) and “failure” (diseased), using both strict and loose criteria defined a priori. Data were recorded in a dedicated chart and updated at every follow-up visit.

All radiographic images were stored in JPEG format, and analyzed using ImageJ software (version 1.41, NIH, Bethesda, MD, USA). To minimize image distortion, the TurboReg plugin (Biomedical Imaging Group, Lausanne, Switzerland) was employed, following the method described by Bose et al. [48]. Two independent, trained, and calibrated evaluators [49], blinded to the treatment outcomes, assigned PAI scores to the preoperative and follow-up radiographs, based on the system proposed by Ørstavik et al. [50]. In cases of disagreement, the higher score was retained. For teeth with multiple roots, the root with the highest PAI score was used for analysis. All the evaluations were performed under standardized viewing conditions, and intra- and inter-examiner agreement was assessed prior to the evaluation phase.

Following PAI score assignment, each tooth was categorized into one of three outcome groups based on combined clinical and radiographic evaluations (Fig. 1):

Fig. 1.

Representative periapical radiographs illustrating treatment outcomes over the follow-up period. Images A–D show a successful case, demonstrating complete resolution and progressive reduction of periapical radiolucency over the 5-year period. Images E–H present a failed case, characterized by initial radiographic improvement at 2 years (G) followed by a relapsing periapical lesion at the 5-year recall (H)

• Healed: the tooth was functional, asymptomatic, and showed no radiographic evidence of apical periodontitis (PAI = 1);

• Healing: the tooth remained functional and asymptomatic but presented with a periapical lesion that had decreased in size (PAI > 1);

• Diseased: the tooth was either non-functional and symptomatic with radiographic signs of apical periodontitis (PAI > 1), or asymptomatic but with an enlarging periapical lesion.

Functional teeth were defined as those in clinical use without any symptoms, regardless of whether apical periodontitis was present or newly developed [51].

Treatment success was defined according to two sets of criteria: under the loose criteria, both healed and healing cases were considered successful, whereas the strict criteria classified only healed cases as successful [6].

Outcome assessment was conducted on the tooth as an entire unit. When the tooth was extracted for complications, the treatment was classified as a failure.

To identify possible prognostic factors several patient-, tooth-, and treatment-related variables were evaluated. Patient factors examined included sex and age of the patient (< 30, 30–55, > 55). Tooth-related factors included initial PAI, tooth position, tooth type, sinus tract, lesion size, apex size, presence of small perforations. Treatment factors evaluated included treatment type (initial treatment or retreatment), sealer extrusion, and number of visits.

Statistical analysis

Statistical analysis was conducted using ad hoc software (STATA BE, version 18.05, StataCorp LP, Texas, United States). Continuous variables were expressed as the Mean with a 95% Confidence Interval (CI); categorical data were tabulated as the number of observations (percentage, %).

Bivariate associations between the treatment outcome and prognostic factors were performed using contingency tables and the Chi-square test or Fisher’s exact test. A multivariate analysis was conducted using logistic regression models to assess the independent association between each prognostic factor and treatment outcome. Two separate models were developed, using as dependent variables the dichotomous outcomes of healing versus diseased based on (1) strict criteria and (2) loose criteria at 2, 3, 4, and 5 years follow-up. The independent variables included treatment type (initial vs. retreatment), absence of extrusion, presence of sinus tract, preoperative PAI score (using PAI 2 as reference), and type of coronal restoration (composite vs. crown). Categorical variables were included as indicator variables using dummy coding.

Results are reported as odds ratios (OR) with 95% confidence intervals (CI) and corresponding p-values. All statistical tests were interpreted at the 5% significance level.

Results

Univariate distribution of prognostic factors

Table 2 presents the demographic and clinical characteristics of the study population. Among the 637 cases included, the majority were female (56.7%), while males accounted for 43.3%. The most represented age category was 30–55 years, comprising nearly half of the sample, followed by individuals older than 55 years and, to a lesser extent, those younger than 30 years.

Table 2.

Descriptive table at tooth level by groups

Prognostic Factor	Categories	Frequenza (%)
Sex	Females	361 (56.67)
	Males	276 (43.33)
	Total	637 (100.00)
Age	< 30	126 (19.78)
	30–55	308 (48.35)
	> 55	203 (31.87)
Tooth Type	Canine	43 (6.75)
	Incisor	135 (21.19)
	Molar	297 (46.62)
	Premolar	162 (25.43)
Arch	Mandible	325 (51.02)
	Maxilla	312 (48.98)
Type of Treatment	Treatment	352 (55.26)
	Retreatment	285 (44.74)
AP Diagnosis	AP C	84 (13.19)
	AP C + RX	496 (77.86)
	AP C + RX+CBCT	57 (8.95)
	Total	637 (100.00)
PAI	1	0 (0.00)
	2	125 (19.62)
	3	104 (16.33)
	4	168 (26.37)
	5	240 (37.68)
Extrusion	Yes	406 (63.74)
	No	231 (36.26)
Apex	> 50	190 (29.83)
	< 50	447 (70.17)
Restoration	Crown	276 (43.33)
	Composite	361 (56.67)
Sinus	Yes	55 (8.63)
	No	582 (91.4)
Dressing	Yes	307 (48.19)
	No	330 (51.81)
Number of visits	Single visit	330 (51.8)
	Multiple visits	307 (48.2)

Molars constituted the most frequently treated tooth type, followed by premolars and incisors. The anatomical distribution was nearly equal between the two arches, with a slightly higher involvement of mandibular teeth compared to maxillary ones.

Analysis of periapical status based on the PAI indicated that no cases presented a score of 1. A minority of teeth exhibited lower PAI scores (2 or 3), whereas the majority showed more severe periapical conditions, with 26.4% scoring 4 and 37.7% scoring 5.

Regarding the type of treatment, 55.3% of cases underwent initial root canal therapy, whereas 44.7% were retreatments. Extrusion was observed in 63.7% of cases, and an apical diameter exceeding 50 units was documented in 29.8% of treatments. In terms of coronal restoration, composite fillings were more commonly placed than full-coverage crowns. Temporary intracanal dressings were applied in 48.2% of the cases, and the presence of a sinus tract was recorded in 55 instances.

Outcome assessed across follow ups

The progressively smaller number of cases available at the 3-, 4-, and 5-year recalls does not reflect patient drop-out. Instead, it is because treatments were performed over a period of time (2018–2023).

From the baseline cohort of 637 cases, all were available at the 1-year and 2-year recalls. 462 cases were examined at 3 years. A total of 323 cases were reviewed at the 4-year recall, and 146 cases at the 5-year recall.

Healing outcomes varied across different follow-up recalls (Table 3). At Recall 1, 80.22% of cases showed healing, while 19% had incomplete healing. A very small percentage of patients experienced no healing (0.31%) or remained symptomatic (0.31%), and vertical root fractures were noted in 0.16% of cases.

Table 3.

Univariate distribution of prognostic factors at follow-ups

Prognostic Factor	Categories	Frequenza (%)
Resorption 4Y	Complete	25 (3.92)
	No	109 (17.11)
	Partial	52 (8.16)
	Total	637 (100.00)
Resorption 5Y	Complete	12 (1.88)
	No	57 (8.95)
	Partial	26 (4.08)
	Total	637 (100.00)
Recall 1	Healed	511 (80.22)
	Healing	121 (19.00)
	Non Healing	2 (0.31)
	Symptomatic	2 (0.31)
	Vertical root fracture	1 (0.16)
	Total	637 (100.00)
Recall 2	Healed	546 (85.71)
	Healing	49 (7.69)
	Non Healing	26 (4.08)
	Perio	2 (0.31)
	Symptomatic	7 (1.10)
	Vertical Root Fracture	7 (1.10)
	Total	637 (100.00)
Recall 3	Healed	401 (86.80)
	Healing	22 (4.76)
	Non Healing	17 (3.68)
	Perio	4 (0.87)
	Symptomatic	7 (1.52)
	Vertical Root Fracture	11 (2.38)
	Total	462 (100.00)
Recall 4	Healed	285 (88.24)
	Healing	8 (2.48)
	Non Healing	14 (4.33)
	Perio	4 (1.24)
	Symptomatic	2 (0.62)
	Vertical Root Fracture	10 (3.10)
	Total	323 (100.00)
Recall 5	Healed	128 (87.67)
	Non healed	4 (2.74)
	Perio	4 (2.74)
	Symptomatic	2 (1.37)
	Vertical Root Fracture	8 (5.48)
	Total	146 (100.00)

At the 2-year recall, 85.7% of cases were classified as healed, while a small proportion showed incomplete healing (7.7%) or no healing (4.1%). Periodontal involvement was rare (0.3%), and both symptomatic cases and vertical root fractures accounted for about 1% each. By 3 years, the number of healed teeth rose slightly to 86.8%, with 4.8% of cases still incompletely healed and 3.7% not healed. Periodontal involvement remained uncommon (0.9%), while symptomatic cases (1.5%) and vertical root fractures (2.4%) were observed in a small minority. At the 4-year follow-up, healing was achieved in 88.2% of cases, whereas 2.5% were still in progress and 4.3% were classified as failures. Periodontal involvement was present in 1.2%, symptoms in 0.6%, and vertical root fractures in 3.1%. At the final 5-year recall, 87.7% of cases showed complete healing, while 2.7% were classified as non-healed and the same proportion presented periodontal involvement. Symptomatic cases remained rare (1.4%), whereas vertical root fractures were observed in 5.5% of cases.

The evaluation of sealer resorption is summarized in Table 3. At the 4-year recall, among the available cases (N = 323), 3.9% exhibited complete resorption, 17.1% showed no evidence of resorption, and 8.2% demonstrated partial resorption.

At the 5-year recall (N = 146), 1.9% showed complete resorption, 9.0% had no resorption, and 4.1% presented partial resorption.

Treatment success over time

Table 4 reports the treatment success rates according to the strict and loose criteria at the 1-, 2-, 3-, 4, and 5-year follow-ups. Using the strict criteria, 80.2% of cases were classified as successful at 1 year, increasing to 85.7% at 2 years, 86.8% at 3 years, 88.2% at 4 years, and 87.67% at 5 years. In contrast, success rates were consistently higher when applying the loose criteria, with 99.2% of cases classified as successful at 1 year, 93.4% at 2 years, 91.6% at 3 years, 90.7% at 4 years, and 87.67% at 5 years.

Table 4.

Strict and Loose Criteria at each follow up

Follow up and Criteria	Success	N (%)
1 year, strict	Yes	511 (80.22)
	No	126 (19.78)
	Total	637 (100)
1 year, loose	Yes	632 (99.22)
	No	5 (0.78)
	Total	637 (100)
2 years, strict	Yes	546 (85.71)
	No	91 (14.29)
	Total	637 (100)
2 years, loose	Yes	595 (93.41)
	No	42 (6.59)
	Total	637 (100)
3 years, strict	Yes	401 (86.80)
	No	61 (13.20)
	Total	462 (100)
3 years, loose	Yes	423 (91.56)
	No	39 (8.44)
	Total	462 (100)
4 years, strict	Yes	285 (88.24)
	No	38 (11.76)
	Total	323
4 years, loose	Yes	293 (90.71)
	No	30 (9.29)
	Total	323
5 years, strict	Yes	128 (87.67)
	No	18 (12.33)
	Total	146 (100)
5 years, loose	Yes	128 (87.67)
	No	18 (12.33)
	Total	146 (100)

The strict criteria consistently identified a greater proportion of failures compared to the loose criteria at all subsequent time points. Despite this difference, both criteria demonstrated a similar overall trend: success rates increased from 1 to 5 years, reaching their highest values at the 4-year recall. At the 5-year recall, no cases were classified as ‘healing’; therefore, strict and loose success rates coincided (Fig. 2).

Fig. 2.

Success rates of endodontic treatment over 1–5 years using strict and loose criteria, showing progressively increasing strict success and consistently high loose success values

Bivariate associations of prognostic factors over different follow-up periods using loose and strict criteria

Over the 5-year follow-up, 203 patients aged > 55 years were evaluated, while 126 patients were aged < 30 years. However, no significant association was found between age and treatment success. Strict success rates remained consistently high across age groups: 66.7–94.7% for < 30, 81.7–85.4% for 30–55, and 80.8–88.2% for > 55, with all p-values > 0.05. These results indicate that advancing age within the cohort did not influence periapical healing outcomes.

At the 1-year evaluation (Table 5), no statistically significant associations were observed between treatment success (according to either loose or strict criteria) and most of the investigated variables, including sex, age group, arch location, type of tooth, PAI score, lesion size, type of treatment (initial vs. retreatment), number of visits, presence of extrusion, apex size, sinus tract, or small perforations.

Table 5.

Bivariate associations between treatment outcome and prognostic factors at 1 year follow up

Prognostic Factor	Category	Success (Loose)		P value	Success (Strict)		P Value
		Yes	NO		Yes	No
Sex	Total	632 (99.2%)	5 (0.8%)	0.880	511 (80.2%)	126 (19.8%)	0.629
	Male	274 (99.3%)	2 (0.7%)		219 (79.3%)	57 (20.7%)
	Female	358 (99.2%)	3 (0.8%)		292 (80.9%)	69 (19.1%)
Age	< 30	124 (98.4%)	2 (1.6%)	0.512	84 (66.7%)	42 (33.3%)	0.06
	30–55	306 (99.4%)	2 (0.6%)		263 (85.4%)	45 (14.6%)
	> 55	202 (99.5%)	1 (0.5%)		164 (80.8%)	39 (19.2%)
Arch	Maxilla	309 (99.0%)	3 (1.0%)	0.621	248 (79.5%)	64 (20.5%)	0.649
	Mandible	323 (99.4%)	2 (0.6%)		263 85.4%)	62 (19.1%)
Tooth Type	Incisor	135 (100.0%)	0 (0.0%)	0.568	104 (77.0%)	31 (23.0%)	0.063
	Canine	43 (100.0%)	0 (0.0%)		29 (67.4%)	14 (32.6%)
	Premolar	160 (98.8%)	2 (1.2%)		130 (80.2%)	32 19.8%)
	Molar	294 (99.0%)	3 (1.0%)		248 (83.5%)	49 (16.5%)
PAI	PAI 2	124 (99.2%)	1 (0.8%)	0.846	108 (86.4%)	17 (13.6%)	0.063
	PAI 3	103 (99.0%)	1 (1.0%)		78 (75.0%)	26 (25.0%)
	PAI 4	166 (98.8%)	2 (1.2%)		140 (83.3%)	28 (16.7%)
	PAI 5	239 (99.6%)	1 (0.4%)		185 (77.1%)	55 (22.9%)
Type of Treatment	Initial	349 (99.2%)	3 (0.8%)	0.831	282 (80.1%)	70 (19.9%)	0.940
	Retreatment	283 (99.3%)	2 (0.7%)		229 (80.4%)	56 (19.6%)
Number of Visits	Single V	328 (99.39%)	2 (0.61%)	0.62	(83.03%)	56 (16.97%)	0.08
	Multiple V	304 (99.02%)	3 (0.98%)		(77.2%)	70 (22.8%)
Extrusion	Yes	404 (99.5%)	2 (0.5%)	0.268	325 (81.0%)	81 (19.0%)	0.886
	No	228 (99.2%)	3 (0.8%)		186 (80.5%)	45 (19.5%)
Apex Size	Apex < 50	443 (99.1%)	4 (0.9%)	0.630	362 (81.0%)	85 (19.0%)	0.457
	Apex > 50	189 (99.5%)	1 (0.5%)		149 (78.4%)	41 (21.6%)
Sinus Tract	Yes	55 (100.0%)	0 (0.0%)	0.490	46 (83.6%)	9 (16.4%)	0.506
	No	577 (99.1%)	5 (0.9%)		465 (79.9%)	117 (20.1%)
Small Perforations	Yes	21 (100.0%)	0 (0.0%)	0.679	17 (81.0%)	4 (19.0%)	0.932
	No	611 (99.2%)	5 (0.8%)		494 (80.2%)	122 (19.8%)
Restorations	Composite	358 (99.2%)	3 (0.8%)	0.880	277 (76.7%)	84 (23.3%)	0.011*
	Crown	274 (99.3%)	2 (0.7%)		234 (84.8%)	42 (15.2%)

However, when applying the strict success criteria, the type of final restoration showed a statistically significant association with outcome (p < 0.05). Specifically, teeth restored with crowns had a higher strict success rate (84.8%) compared to those restored with composite restorations (76.7%).

Overall, the data suggest that restoration type may influence clinical outcomes at 1 year, particularly when using more stringent evaluation criteria.

At the 2-year evaluation (Table 6), none of the assessed prognostic factors showed a statistically significant association with treatment success when applying either the loose or strict criteria (all p > 0.05).

Table 6.

Bivariate associations between treatment outcome and prognostic factors at 2 years follow up

Prognostic Factor	Category	Success Loose		P value	Success (Strict)		P Value
		Yes	No		Yes	No
Sex	Total	632 (99.2%)	5 (0.8%)	0.88	511 (80.2%)	126 (19.8%)	0.629
	Male	274 (99.3%)	2 (0.7%)		219 (79.3%)	57 (20.7%)
	Female	358 (99.2%)	3 (0.8%)		292 (80.9%)	69 (19.01%)
Age	< 30	287 (93.2%)	21 (6.8%)	0.63	268 (87.0%)	40 (13.0%)	0.353
	30–55	120 (95.2%)	6 (4.8%)		103 (81.7%)	23 (18.3%)
	> 55	188 (92.6%)	15 (7.4%)		175 (86.2%)	28 (13.8%)
Arch	Maxilla	304 (93.5%)	21 (6.5%)	0.89	280 (86.2%)	45 (13.8%)	0.746
	Mandible	291 (93.5%)	21 (6.5%)		266 (85.3%)	46 (14.7%)
Tooth Type	Incisor	127 (94.1%)	8 (5.9%)	0.99	111 (82.2%)	24 (17.8%)	0.063
	Canine	40 (93.0%)	3 (7%)		35 (81.4%)	8 (18.6%)
	Premolar	151 (93.0%)	11 (6.8%)		138 (85.2%)	24 (14.8%)
	Molar	277 (93.3%)	20 (6.7%)		262 (88.2%)	35 (11.8%)
PAI	PAI 2	119 (95.2%)	6 (4.8%)	0.76	119 (95.2%)	6 (4.8%)	0.063
	PAI 3	98 (94.2%)	6 (5.8 %)		93 (89.4%)	11 (10.6%)
	PAI 4	156 (92.9%)	12 (7.1%)		148 (88.1%)	20 (11.9%)
	PAI 5	222 (92.2%	18 (7.5%)		186 (77.5%)	54 (22.5%)
Type of Treatment	Initial	333 (94.6%)	19 (5.4%)	0.18	304 (86.4%)	48 (13.6%)	0.940
	Retreatment	262 (94.6%)	23 (5.4%)		242 (86.4%)	43 (13.6%)
Number of Visits	Single V	287 (93.3%)	20 (6.7%)	0.94	257 (87.6%)	50 (12.4%)
	Multiple V	308 (93.3)	22 (6.7%)		289 (87.6%)	41 (12.4%)	0.164
Extrusion	Yes	381 93.8%)	25 (6.2%)	0.56	344 (84.7%)	62 (15.3%)	0.346
	No	214 (92.6%)	17 (7.4%)		202 (31.7%)	29 (4.6%)
Apex Size	< 50	418 (93.5%)	29 (6.5%)	0.87	384 (85.9)	63 (14.1%)	0.832
	> 50	177 (93.5%)	13 (6.8%)		162 (86.0%)	28 (14.0%)
Sinus Tract	Yes	546 ( 93.8%)	36 (6.2%)	0.18	499 (85.7%)	83 (14.3%)	0.954
	No	49 (89.1%)	6 (10.9%)		47 (85.5%)	8 (14.5%)
Small Perforations	Yes	19 (90.5%)	2 (9.5%)	0.58	16 (76.2%)	5 (23.8%)	0.205
	No	576 (93.5%)	40 (6.5%)		530 (86.0%)	86 (14.0%)
Restorations	Composite	341 (94.5%)	20 (5.5%)	0.22	302 (83.7%)	59 (16.3%)	0.090
	Crown	254 (92.0%)	22 (8.0%)		244 (88.4%)	32 (11.6%)

With the strict criteria, teeth with higher PAI scores at baseline tended to show lower success rates, particularly those with a score of 4 (88.1% success) and 5 (77.5%), compared to teeth with PAI 2 (95.2%) or 3 (89.4%), without showing statistical significance.

Similarly, tooth type showed a non-significant trend (p > 0.05), where molars had a slightly lower strict success rate (88 .2%) compared to premolars (85.2%) and incisors (82.2%).

The type of restoration did not show statistical significance for strict success (p > 0.05), with crowns yielding a slightly higher success rate (88.4%) compared to composite restorations (83.7%).

Other variables, including sex, age, arch, lesion size, treatment type, number of visits, extrusion, apex size, sinus tract, and presence of small perforations, were not significantly associated with treatment outcome at 2 years (p > 0.05 for all).

At the 3-year evaluation (Table 7), most prognostic factors were not significantly associated with treatment success when assessed with either the loose or strict criteria (all p > 0.05). Variables such as sex, age, arch location, baseline PAI, lesion size, treatment type, number of visits, extrusion, apex size, sinus tract, and type of restoration did not show significant associations. However, when applying the strict criteria, small perforations were significantly associated with reduced success (p<0.05), with perforated teeth exhibiting lower healing rates compared to non-perforated teeth. Tooth type did not significantly affect success rates (p > 0.05), with molars showing slightly lower strict success compared to premolars and incisors.

Table 7.

Bivariate associations between treatment outcome and prognostic factors at 3 years follow up

Prognostic Factor	Category	Success (Loose)		P value	Success (Strict)		P Value
		Yes	No		Yes	No
Sex	Male	190 (91.8%)	17 (8.2%)	0.873	180 (87.0%)	27 (13.0%)	0.927
	Female	233 (91.4%)	22 (8.6%)		221 (86.7%)	34 (13.3%)
Age	< 30	95 (91.3%)	9 (8.7%)	0.989	90 (86.5%)	14 (13.5%)	0.992
	30–55	193 (91.5%)	18 (8.5%)		183 (86.9%)	28 (13.1%)
	> 55	135 (91.2%)	12 (8.8%)		128 (87.1%)	19 (12.9%)
Arch	Maxilla	215 (91.5%)	20 (8.5%)	0.621	208 (88.5%)	27 (11.5%)	0.168
	Mandible	208 (91.6%)	19 (8.4%)		193 (85.0%)	34 (15.0%)
Tooth Type	Incisor	101 (92.1%)	8 (7.3%)	0.774	111 (82.2%)	24 (17.8%)	0.063
	Canine	25 (96.2%)	1 (3.8%)		35 (94.6%)	2 (5.4%)
	Premolar	120 (90.2%)	12 (9.8%)		95 (88.8%)	12 (11.2%)
	Molar	177 (93.6%)	12 (6.4%)		135 (85.8%)	22 (14.2%)
PAI	PAI 2	56 (94.9%)	3 (5.1%)	0.539	56 (94.9%)	3 (5.1%)	0.104
	PAI 3	51 (87.9%)	7 (12.1%)		47 (81.0%)	11 (19.0%)
	PAI 4	137 (92.6%)	11 (7.4%)		131 (88.5%)	17 (11.5%)
	PAI 5	179 (90.9%)	18 (9.1%)		167 (84.8%)	30 (15.2%)
Type of Treatment	Initial	209 (92.9%)	16 (7.1%)	0.316	199 (88.2%)	26 (11.8%)	0.940
	Retreatment	214 (90.3%)	23 (9.7%)		202 (85.2%)	35 (14.8%)
Number of Visits	Single	207 (90.8%)	21 (9.2%)	0.557	198 (86.8%)	30 (13.2%)	0.977
	Multiple	206 (91.1%)	18 (8.0%)		203 (86.5%)	31 (13.5%)
Extrusion	Yes	277 (92.5%)	22 (7.5%)	0.256	256 (85.0%)	45 (15.0%)	0.487
	No	146 (89.6%)	17 (10.4%)		140 (83.8%)	27 (16.2%)
Apex Size	Apex < 50	281 (93.1%)	21 (6.9%)	0.114	266 (88.1%)	36 (11.9%)	0.263
	Apex > 50	142 (88.7%)	18 (11.3%)		135 (84.4%)	25 (15.6%)
Sinus Tract	Yes	36 (84.4%)	7 (15.6%)	0.397	35 (83.3%)	7 (16.7%)	0. 487
	No	386 (92.9%)	34 (7.1%)		366 (87.1%)	54 (12.9%)
Perforations	Yes	13 (86.7%)	2 (13.3%)	0.679	10 (66.7%)	5 (33.3%)	0.019*
	No	410 (91.7%)	37 (8.3%)		391 (87.5%)	56 (12.5%)
Restorations	Composite	244 (90.7%)	25 (9.3%)	0.437	229 (85.1%)	40 (14.9%)	0.212
	Crown	179 (92.8%)	14 (7.2%)		172 (89.1%)	21 (10.9%)

At the 4-year evaluation (Table 8), some variables showed significant associations with outcome. Using the loose criteria, teeth with smaller apical diameters (< 50) had higher success rates compared to those with larger apices (> 50) (p < 0.05). Under the strict criteria, baseline PAI was significantly associated with success (p < 0.05), with teeth presenting lower PAI scores showing higher healing rates. Similarly, apex size was significant (p < 0.05), with smaller apices linked to better outcomes. In addition, small perforations were significantly associated with strict success (p < 0.05). Other factors, including sex, age, arch location, tooth type, treatment type, dressing, sinus tract, restoration type, extrusion, and resorption, were not significantly related to outcome (p > 0.05).

Table 8.

Bivariate associations between treatment outcome and prognostic factors at 4 years follow up

Prognostic Factor	Category	Success Loose		P Value Loose	Success Strict		P Value Strict
		Yes	No		Yes	No
Sex	M	142 (91.6%)	13 (8.4%)	0.592	139 (89.7%)	16 (10.3%)	0.440
	F	151 (89.9%)	17 (10.1%)		146 (86.9%)	22 (13.1%)
Age	< 30	73 (94.8%)	4 (5.2%)	0.283	73 (94.8%)	4 (5.2%)	0.083
	30–55	120 (88.2%)	16 (11.8%)		115 (84.6%)	21 (15.4%)
	> 55	100 (88.5%)	10 (11.5%)		97 (88.2%)	13 (11.8%)
Location	Mandible	144 (91.1%)	14 (8.9%)	0.796	138 (87.3%)	20 (12.7%)	0.626
	Maxilla	149 (90.3%)	16 (9.7%)		147 (89.1%)	18 (10.9%)
ToothType	Canine	16 (94.1%)	1 (5.9%)	0.965	14 (82.4%)	3 (17.6%)	0.547
	Incisor	64 (90.1%)	7 (9.9%)		60 (84.5%)	11 (15.5%)
	Molar	128 (90.8%)	13 (9.2%)		126 (89.4%)	15 (10.6%)
	Premolar	85 (90.4%)	9 (9.6%)		85 (90.4%)	9 (9.6%)
PAI	2	46 (95.8%)	2 (4.2%)	0.332	46 (95.8%)	2 (4.2%)	0.011*
	3	46 (90.2%)	5 (9.8%)		46 (90.2%)	5 (9.8%)
	4	97 (92.4%)	8 (7.6%)		97 (92.4%)	8 (7.6%)
	5	104 (87.4%)	15 (12.6%)		96 (80.0%)	23 (20.0%)
Treatment	Initial	140 (92.7%)	11 (7.3%)	0.245	138 (84.7%)	13 (15.3%)	0.099
	Retreatment	153 (88.9%)	19 (11.1%)		147 (85.5%)	25 (14.5%)
Apex	< 50	196 (93.3%)	14 (6.7%)	0.027*	191 (91.0%)	19 (16.1%)	0.039*
	> 50	97 (85.8%)	16 (14.2%)		94 (83.2%)	19 (9.0%)
Perforation	Yes	12 (85.7%)	2 (14.3%)	0.510	10 (71.4%)	4 (28.6%)	0.046*
	No	281 (90.9%)	28 (9.1%)		275 (89.0%)	34 (11.0%)
Dressing	Yes	148 (89.2%)	18 (10.8%)	0.322	146 (87.9%)	20 (12.1%)	0.871
	No	145 (92.4%)	12 (7.6%)		139 (88.5%)	18 (11.5%)
Sinus	Yes	267 (91.4%)	25 (8.6%)	0.168	259 (88.7%)	33 (11.3%)	0.428
	No	26 (83.9%)	5 (16.1%)		26 (83.9%)	5 (16.1%)
Restoration	Composite	167 (90.3%)	18 (9.7%)	0.751	163 (88.1%)	22 (11.9%)	0.935
	Crown	126 (91.3%)	12 (8.7%)		122 (88.4%)	16 (11.6%)
Extrusion	Yes	169 (90.9%)	17 (9.1%)		164 (88.2%)	22 (11.8%)	0.967
	No	124 (92.0%)	11 (8.0%)	0.915	121 (88.3%)	16 (11.7%)
Resorption	NR	133 (87.5%)	19 (12.5%)	0.065	130 (85.5%)	22 (14.5%)	0.157
	Complete	18 (90.0%)	2 (10.0%)		18 (90.0%)	2 (10.0%)
	None	96 (96.0%)	4 (4.0%)		95 (95.0%)	5 (5.0%)
	Partial	46 (90.2%)	5 (9.8%)		42 (82.4%)	9 (17.6%)

Nr, not reported

At the 5-year evaluation (Table 9), most variables remained unrelated to treatment success under either set of criteria. Sex, age, arch, tooth type, baseline PAI, treatment type, apex size, small perforations, dressing, sinus tract, restoration type, extrusion, and resorption were not significantly associated with outcome.

Table 9.

Bivariate associations between treatment outcome and prognostic factors at 5 years follow up

Prognostic Factor	Category	Success Loose		P Value Loose	Success Strict			P Value Strict
		Yes	No		Yes	No
SEX	M	58 (89.2%)	7 (10.8%)	0.608	58 (89.2%)		7 (10.8%)	0.608
	F	70 (86.4%)	11 (13.6%)		70 (86.4%)		11 (13.6%)
AGE	< 30	32 (91.4%)	3 (8.6%)	0.718	32 (91.4%)		3 (8.6%)	0.718
	30–55	48 (85.7%)	8 (14.3%)		48 (85.7%)		8 (14.3%)
	> 55	48 (87.3%)	7 (12.7%)		48 (87.3%)		7 (12.7%)
Location	Mandible	61 (89.7%)	7 (10.3%)	0.485	61 (89.7%)		7 (10.3%)	0.485
	Maxilla	67 (85.9%)	11 (14.1%)		67 (85.9%)		11 (14.1%)
ToothType	Canine	9 (90.0%)	1 (10.0%)	0.964	9 (90.0%)		1 (10.0%)	0.964
	Incisor	27 (90.0%)	3 (10.0%)		27 (90.0%)		3 (10.0%)
	Molar	53 (86.9%)	8 (13.1%)		53 (86.9%)		8 (13.1%)
	Premolar	39 (86.7%)	6 (13.3%)		39 (86.7%)		6 (13.3%)
PAI	2	19 (90.5%)	2 (9.5%)	0.902	19 (90.5%)		2 (9.5%)	0.902
	3	21 (84.0%)	4 (16.0%)		21 (84.0%)		4 (16.0%)
	4	41 (89.1%)	5 (10.9%)		41 (89.1%)		5 (10.9%)
	5	47 (87.0%)	7 (13.0%)		47 (87.0%)		7 (13.0%)
Treatment	Initial	61 (84.7%)	7 (10.3%)	0.485	61 (84.7%)		7 (10.3%)	0.485
	Retreatment	67 (85.9%)	11 (14.1%)		67 (85.9%)		11 (14.1%)
Apex	< 50	81 (90.0%)	9 (10.0%)	0.278	81 (90.0%)		9 (10.0%)	0.278
	> 50	47 (84.0%)	9 (16.0%)		47 (84.0%)		9 (16.0%)
Small Perforation	Yes	3 (60.0%)	2 (40.0%)	0.055	3 (60.0%)		2 (40.0%)	0.055
	No	125 (88.6%)	16 (11.4%)		125 (88.6%)		16 (11.4%)
Dressing	Yes	65 (86.7%)	10 (13.3%)	0.704	65 (86.7%)		10 (13.3%)	0.704
	No	63 (88.7%)	8 (11.3%)		63 (88.7%)		8 (11.3%)
Sinus	Yes	111 (87.4%)	16 (12.6%)	0.798	111 (87.4%)		16 (12.6%)	0.798
	No	17 (89.5%)	2 (10.5%)		17 (89.5%)		2 (10.5%)
Restoration	Composite	71 (87.7%)	10 (12.3%)	0.994	71 (87.7%)		10 (12.3%)	0.994
	Crown	57 (87.7%)	8 (12.3%)		57 (87.7%)		8 (12.3%)
Extrusion	Yes	87 (91.6%)	8 (8.4%)	0.05	87 (91.6%)		8 (8.4%)	0.05
	No	41 (80.4%)	10 (19.6%)		41 (80.4%)		10 (19.6%)
Resorptions	NR	41 (87.2%)	10 (12.8%)	0.194	41 (87.2%)		10 (12.8%)	0.194
	Complete	10 (83.3%)	2 (16.7%)		10 (83.3%)		2 (16.7%)
	None	53 (93.0%)	4 (7.0%)		53 (93.0%)		4 (7.0%)
	Partial	24 (92.3%)	2 (7.7%)		24 (92.3%)		2 (7.7%)

Multivariate logistic regression

Multivariate logistic regression analysis was performed to identify prognostic factors associated with endodontic treatment success at 2, 3, 4, 5 according to both strict and loose outcome criteria (Table 10).

Table 10.

Multivariate logistic regression analysis evaluating the association between prognostic factors and treatment success at 1, 2, 3, 4, and 5 years according to strict and loose criteria. Odds ratios (OR) with 95% confidence intervals (CI) and p-values are reported

Variable	Comparison (vs. reference)	Strict Criteria OR (95% CI)	p	Loose Criteria OR (95% CI)	p
1 year
Treatment	Retreatment (vs. Initial)	0.96 (0.63–1.47)	0.87	0.4 (0.06–2.53)	0.331
Extrusion	No (vs. Yes)	0.96 (0.64–1.48)	0.91	0.74 (0.12–5.04)	0.76
Sinus Tract	Yes (vs. No)	0.77 (0.36–1.62)	0.49	1.05 (0.67–1.63)	0.81
PAI	3 (vs. 2)	0.65 (0.42–0.98)	0.02*	0.70 (0.44–1.10)	0.74
PAI	4 (vs. 2)	0.50 (0.27–0.93)	0.034*	0.62 (0.33–1.16)	0.55
PAI	5 (vs. 2)	0.38 (0.18–0.81)	0.03*	0.84 (0.05–15.52)	0.91
Restoration	Crown (vs. Composite)	0.59 (0.39–0.9)	0.014*	0.88 (0.14–5.37)	0.89
2 years
Treatment	Retreatment (vs. Initial)	0.97 (0.60–1.56)	0.889	0.72 (0.37–1.42)	0.347
Extrusion	No (vs. Yes)	1.03 (0.63–1.69)	0.914	1.40 (0.72–2.72)	0.316
Sinus Tract	Yes (vs. No)	1.09 (0.49–2.44)	0.832	0.57 (0.23–1.45)	0.241
PAI	3 (vs. 2)	0.41 (0.14–1.18)	0.098	0.94 (0.28–3.14)	0.925
PAI	4 (vs. 2)	0.35 (0.13–0.94)	0.038*	0.76 (0.26–2.26)	0.626
PAI	5 (vs. 2)	0.17 (0.07–0.42)	0.000*	0.66 (0.24–1.82)	0.426
Restoration	Crown (vs. Composite)	1.50 (0.93–2.43)	0.097	0.72 (0.38–1.37)	0.319
3 years
Treatment	Retreatment (vs. Initial)	1.33 (0.75–2.75-36)	0.33	1.43 (0.711–2.88)	0.32
Extrusion	No (vs. Yes)	0.69 (0.39–1.2)	0.2	0.65 (0.33–1.27)	0.21
Sinus Tract	Yes (vs. No)	1.40 (0.58–3.37)	0.45	1.58 (0.58–4.36)	0.37
PAI	3 (vs. 2)	0.62 (0.40–0.97)	0.04*	0.71 (0.44–1.15)	0.2
PAI	4 (vs. 2)	0.50 (0.27–0.93)	0.03*	0.63 (0.33–1.20)	0.39
PAI	5 (vs. 2)	0.44 (0.21–0.91)	0.03*	0.56 (0.26–1.18)	0.55
Restoration	Crown (vs. Composite)	0.65 (0.36–1.17)	0.15	0.71(0.35–1.44)	0.34
4 years
Treatment	Retreatment (vs. Initial)	1.68 (0.79–3.59)	0.177	1.41 (0.62–3.2)	0.41
Extrusion	No (vs. Yes)	0.85 (0.41–1.75)	0.66	0.81 (0.37–1.77)	0.59
Sinus Tract	Yes (vs. No)	1.42 (0.49–4.15)	0.521	2.07 (0.70–6.09)	0.19
PAI	3 (vs. 2)	0.70 (0.37–1.32)	0.424	0.76 (0.39–1.48)	0.37
PAI	4 (vs. 2)	0.52 (0.28–0.98)	0.62	0.61 (0.31–1.19)	0.52
PAI	5 (vs. 2)	0.40 (0.19–0.85)	0.04*	0.48 (0.22–1.05)	0.16
Restoration	Crown (vs. Composite)	0.922 (0.45–1.89)	0.83	0.82 (0.37–1.80)	0.62
5 years
Treatment	Retreatment (vs. Initial)	0.94 (0.45–1.95)	0.6	0.97 (0.47–2.01)	0.59
Extrusion	No (vs. Yes)	0.34 (0.12–0.98)	0.05	0.34 (0.12–0.98)	0.05
Sinus Tract	Yes (vs. No)	0.88 (0.35–2.23)	0.85	0.91 (0.37–2.27)	0.85
PAI	3 (vs. 2)	0.76 (0.30–1.95)	0.541	0.82 (0.33–2.05)	0.54
PAI	4 (vs. 2)	0.65 (0.25–1.71)	0.915	1.11 (0.17–7.42)	0.92
PAI	5 (vs. 2)	0.58 (0.22–1.55)	0.58		0.58
Restoration	Crown (vs. Composite)	0.97 (0.35–2.67)	0.94	0.97 (0.35–2.67)	0.95

*p value < 0.05

1-year follow up

At the 1-year recall, under strict criteria, the model was statistically significant. Baseline PAI score was a strong prognostic indicator: compared with PAI 2, teeth with PAI 3 (OR = 0.65; 95% CI: 0.42–0.98; p = 0.02), PAI 4 (OR = 0.50; 95% CI: 0.27–0.93; p = 0.034) and PAI 5 (OR = 0.38; 95% CI: 0.18–0.81; p = 0.03) had significantly lower odds of success. Coronal restoration was also significant: crowns were associated with lower odds of success compared to composite restorations (OR = 0.59; 95% CI: 0.39–0.90; p = 0.014). Other factors, including retreatment vs. initial treatment, extrusion, and sinus tract, were not significantly associated with outcome.

Under loose criteria, the overall model was not significant, and none of the predictors showed a significant effect.

2-year follow up

At the 2-year recall, higher baseline PAI remained significantly associated with reduced success under strict criteria. Compared with PAI 2, PAI 4 (OR = 0.35; 95% CI: 0.13–0.94; p = 0.038) and PAI 5 (OR = 0.17; 95% CI: 0.07–0.42; p < 0.001) were strong negative prognostic factors, while PAI 3 showed a similar but non-significant trend (OR = 0.41; 95% CI: 0.14–1.18; p = 0.098). None of the other variables (treatment type, extrusion, sinus tract, or restoration) were significantly associated with outcome.

Under loose criteria, the model was not statistically significant, and none of the predictors reached significance.

3-year follow up

At the 3-year recall, higher baseline PAI scores were significantly associated with reduced odds of success under strict criteria. Compared with PAI 2, teeth with PAI 3 (OR = 0.62; 95% CI: 0.40–0.97; p = 0.04), PAI 4 (OR = 0.50; 95% CI: 0.27–0.93; p = 0.03), and PAI 5 (OR = 0.44; 95% CI: 0.21–0.91; p = 0.03) showed significantly lower success rates.

Other variables, including treatment type, extrusion, sinus tract, and restoration, were not significantly associated with outcome.

Under loose criteria, none of the variables reached statistical significance.

4-year follow up

At the 4-year recall, baseline PAI 5 was the only significant prognostic factor, with teeth showing a markedly lower probability of success compared with PAI 2 (OR = 0.40; 95% CI: 0.19–0.85; p = 0.04). PAI 3 and PAI 4 showed similar trends but did not reach significance. Other factors, including retreatment vs. initial treatment, extrusion, sinus tract, and restoration, were not significantly associated with outcome.

Under loose criteria, no significant associations were observed.

These findings indicate that at 4 years, a higher PAI score (PAI 5) was the only prognostic factor significantly associated with reduced success under strict criteria, while no associations were found when using loose criteria.

5-year follow up

At 5-years follow-up, baseline PAI scores (3, 4, and 5 vs. 2) showed a consistent trend toward lower success, but none of the associations reached statistical significance. Similarly, treatment type, sinus tract, extrusion, and restoration were not significantly related to treatment outcome. However, interestingly, the absence of extrusion was associated with lower odds of success compared to its presence under strict (OR = 0.33; 95% CI: 0.11–0.98; p = 0.05) and loose criteria (OR = 0.34; 95% CI: 0.12–0.98; p = 0.05). Although this finding was not statistically significant.

Overall, these findings suggest that at 5 years, no significant effects were observed for clinical or procedural variables.

Sealer Extrusion and Resorption Rates

Extrusions and resorption rates are shown in Table 11. At the 5-year follow-up, sealer extrusion was observed in 186 cases, with an overall healing rate of 90.9%. Among these, the majority of cases showed no resorption (58.6%), followed by partial (28%) and complete resorption (13.4%). Teeth without sealer resorption exhibited the highest healing rate (92.7%), whereas slightly lower success rates were observed in cases with partial (88.5%) and complete resorption (88.0%).

Table 11.

Analysis of unintentional extrusion and resorption rates (5 years)

Category	TOTAL N	TOTAL %	HEALED N	HEALED %	NOT HEALED N	NOT HEALED %
Extrusion	186	100	169	90.9	17	9.1
No Resorption	109	58.6	101	92.7	8	7.3.0
Partial Resorption	52	28.0	46	88.5	6	11.5
Complete Resorption	25	13.4	22	88.0	3	12.0

Discussion

This retrospective cohort study aimed to evaluate the clinical efficacy of CSBSs with CHC in teeth with apical periodontitis and to identify patient-, tooth-, and treatment-related factors associated with success or failure. Within the limitations of this study, the overall success rate of nonsurgical endodontic treatment and retreatment in a private practice setting using CSBSs and a CHC was 87.67% at 5 years of follow-up.

The success rate observed in the present study compares favorably with the existing literature on non-surgical endodontic treatment of teeth with apical periodontitis. Previous reports have shown variable outcomes, with success rates ranging from approximately 75% to 89%, depending on study design, follow-up time, and case selection [13, 52–54]. For example, Moazami et al., reported 87.5% success rate in 104 permanent teeth at 7 years; the rate of complete healing for teeth with initial treatment was 89.7%, and for retreatment group was 85.7% [52]. Peters et al. have had a success rate of about 75.5% in 102 teeth with mean interval of 25.4 ± 11.8 months; the healing rate for teeth with initial treatment was 85%, and for retreatment group was 61.2% [53]. Furthermore, in the Toronto study, the 4- to 6-year success rate reported for teeth with apical periodontitis was 82% in initial treatment cases and 80% in retreatment cases [13, 54].

Other investigations have yielded similar results, such as Ørstavik et al. [11] (≈ 79% at 6 months–4 years) and Ricucci et al. (82.7% at 5 years) [55]. In addition, data from a previous meta-analysis demonstrated that the presence of apical periodontitis lowers prognosis, with success rates of about 82% in teeth without lesions and 72% in those with lesions [56]. In a large sample size study, the reported success rate was 89.1% for the initial root canal treatment and 85.6% for retreatment [6]. Taken together, these data suggest that the outcomes achieved in the present study are at least comparable, and in some cases superior, to those previously reported in similar clinical contexts.

Few clinical studies are available on the success rates of non-surgical root canal treatments and/or retreatments using premixed CSBS sealers, specifically CeraSeal. While most of the aforementioned studies compared CSBSs with conventional sealers, fewer clinical investigations have specifically assessed the performance of premixed CSBSs [57, 58]. Furthermore, limited long-term evidence is available [30].

The use of CeraSeal in this study demonstrated favorable clinical behavior, confirming its potential as a viable obturation material. In terms of physicochemical properties, it was observed that CeraSeal maintains an alkaline pH, releases substantial calcium ions, and demonstrates low solubility and high radiopacity, meeting ISO standards for endodontic sealers [33, 59, 60]. Its porosity and hydrophilicity are moderate, and its solubility is lower than that of some other bioceramic sealers, which may contribute to its durability [33, 59, 60].

The present study findings align with those shown by a previous study, which reported an 85.4% success rate using CeraSeal and CHC technique after three years [38]. These data highlight the progressive nature of periapical healing and support the long-term effectiveness of CeraSeal in routine endodontic practice. However, these data differ slightly from those of another study, which reported a success rate of endodontic treatment with CSBS, including CeraSeal, of about 99.0% at a mean follow-up of approximately 20 months when used with the warm gutta-percha obturation technique. In that study, both healed and healing cases were considered as successes, and no statistically significant differences were found among the tested sealers or between teeth with and without periapical lesions. The reported success rate was 100% for initial treatments and 98.2% for retreatments [61].

In recent clinical studies and meta-analyses, the percentage of success of endodontic treatment using various CSBS in teeth with apical periodontitis is consistently high. Prospective and retrospective clinical studies report success rates ranging from 90.9% to 99% at follow-up periods of 1–2 years [34, 61–63]. A recent meta-analysis reports a pooled failure rate of 6.8% for nonsurgical endodontic treatment and retreatment using CSBSs with CHC technique, which translates to a success rate of approximately 93.2% in teeth with apical periodontitis [64]. Most evidence reports results with less than 2 years of follow-up; the aim of this study is to enhance those findings by providing data from a much larger sample (n = 637), observed over up to 5 years, with a consistent operator protocol. This helps reduce confounding factors related to operator variability and technique differences.

Healing outcomes improved over five years, with healed cases increasing steadily under strict criteria. Loose success remained high but slightly declined after the first recall. Strict criteria are more sensitive to minor radiographic changes, while loose criteria may mask partial failures [34]. Compared to previous data, results show similar trends with higher strict success rates. In the present study, loose criteria stayed high (99.2% at 1 year, 93.4% at 2 years, 91.6% at 3 years, 90.7% at 4 years), but strict success was always higher, rising from 80.2% at 1 year to 88.2% at 4 years. On the contrary, previous studies showed short-term follow-ups (6–18 months) and report 93–99% success by loose criteria and 51–73% by strict criteria, with the gap narrowing at medium term (90–99% vs. 60–76%) and persisting long term (84–87% vs. 76–78%). The present study findings suggest that the use of CeraSeal with a standardized CHC technique may favor faster and more complete radiographic healing compared to previously reported outcomes [18, 30, 61].

One of the most relevant prognostic indicators was the baseline PAI score. While bivariate analysis did not consistently show a statistically significant association, multivariate analysis demonstrated that higher PAI scores were associated with reduced success, particularly when strict criteria were applied. This suggests that the effect of baseline lesion severity becomes more evident after adjustment for potential confounding factors.

At the 1-year follow-up, baseline PAI emerged as a significant predictor of outcome in the multivariate model, with progressively lower odds of success for increasing PAI categories. At 2 years, PAI 4 and 5 were significantly associated with lower odds of success, while PAI 3 showed a similar but non-significant trend. At 3 years, higher PAI scores remained associated with reduced success, with PAI 3, PAI 4, PAI 5 showing a significantly increased risk of failure. At 4 years, only PAI 5 remained significantly associated with reduced strict success, whereas at the 5-year follow-up, no statistically significant association was observed. This lack of correlation may reflect the smaller sample size available at the longest follow-up, but it may also suggest that, given sufficient time, even cases with initially severe lesions can eventually achieve healing. These findings are in line with previous studies confirming that the initial severity of periapical disease is the strongest predictor of treatment outcome [61, 65, 66]. Specifically, an earlier investigation demonstrated that teeth with periapical lesions < 5 mm had a success rate of 86.6%, whereas in cases with lesions > 5 mm, the success rate was 78.2% [55].

The relationship between baseline PAI scores and treatment outcome appears to be time-dependent, as baseline lesion severity was already a significant predictor at early follow-ups, particularly in multivariate analysis, and remained evident at intermediate time points. At longer follow-ups, the association became less consistent, which may be partly explained by the reduced sample size and consequent lower statistical power. In addition, the progressive nature of periapical healing may contribute to this pattern, as even teeth with initially severe lesions can achieve radiographic resolution over time. Therefore, the prognostic influence of baseline PAI scores appears to be more evident in the early and intermediate phases of healing, while becoming less pronounced at longer follow-ups.

The relatively high rate of sealer extrusion observed in the present study (63.7%) may be explained by the characteristics of CSBSs and the CHC technique, in accordance with a previous study showing 66% of extrusion [34]. The high flowability and hydrophilicity of these materials, combined with the hydraulic pressure generated during cone insertion, may facilitate sealer displacement beyond the apical foramen, particularly in teeth with apical periodontitis.

However, extrusion was not associated with worse outcomes. This finding is consistent with the known biocompatibility of CSBSs. The lack of a significant effect of sealer extrusion across follow-ups may be attributed to the high biocompatibility of CSBSs, which are generally well tolerated by periapical tissues [34, 67]. Interestingly, a slightly higher success rate was observed in teeth presenting sealer extrusion compared with those without extrusion. Although this difference did not reach statistical significance in multivariate analysis, it may further support the favorable biological response to extruded calcium silicate–based materials.

The sealer resorption observed in the present study may be related to the intrinsic physicochemical properties of CSBSs [68], which demonstrate significantly higher solubility than epoxy resin-based sealers, with values often exceeding the ISO 6876 standard of 3% mass loss [69, 70]. This behavior is closely linked to their bioactivity, as the release of calcium ions promotes the formation of apatite-like deposits and interaction with periapical tissues. This solubility is not simply material degradation but reflects the dynamic ion exchange and hydration processes inherent to these sealers [71]. Recent data show that Ceraseal maintains relatively stable crystalline structures over extended periods (up to 6 months) and has moderate solubility compared to other CSBSs [69].

From a chemical perspective, CSBSs set through hydration reactions, leading to the formation of calcium silicate hydrate (C–S–H) gel and calcium hydroxide. While the C–S–H phase provides structural stability, calcium hydroxide contributes to alkalinity and ion release. In some formulations, calcium aluminate phases may also be present and can influence the kinetics of hydration and early setting reactions. These mechanisms may contribute to the gradual resorption patterns observed over time [72]. Therefore, the partial resorption of extruded material should not necessarily be interpreted as a negative finding, but rather as a reflection of the material’s interaction with the biological environment. However, the potential impact of this process on long-term sealing ability remains to be further elucidated.

Among other factors, small iatrogenic perforations were associated with worse outcomes at intermediate recalls. At 3 years, perforated teeth had significantly lower strict success (p < 0.05), and this finding persisted at 4 years (p < 0.05). These results emphasize the adverse impact of iatrogenic events on periapical healing. Moreover, at 4 years, apical anatomy emerged as a significant determinant of outcome. Smaller apical diameters (< 50) were associated with higher success rates under both loose (p < 0.05) and strict criteria (p < 0.05), suggesting that large apices may compromise obturation quality and sealing ability. At the 5-year recall, however, neither small perforations nor apical diameter showed a significant influence on treatment outcome. This again may in part reflect the reduced sample size available at the longest follow-up, but it could also suggest that, over extended healing periods, the negative impact of procedural complications and anatomical variations becomes less pronounced compared with earlier time points. Additionally, the perforations included in this study were iatrogenic, occurring during instrumentation with manual or rotary files, and were therefore generally small and more likely located in the middle or apical thirds of the root canal. However, detailed and standardized information on their exact timing and size was unavailable, limiting a more in-depth analysis of this variable.

The inclusion of teeth with large apical diameters (> ISO size 50) may raise concerns regarding the indication for alternative procedures such as apexification. However, all teeth included in the present study had complete root development and were diagnosed with apical periodontitis, for which conventional orthograde treatment remains the standard approach [73]. In this context, CSBSs may offer specific advantages, as their bioactive properties, flowability, and ability to form mineralized tissue may facilitate apical sealing even in wide foramina. The CHC technique, which relies on sealer adaptation rather than compaction forces, may further support this approach [73].

The progressive aging of the cohort did not translate into differences in outcome. Strict success rates remained high across all age groups at each recall, with no significant associations (all p > 0.05). Similarly, sex, arch location, and tooth type did not significantly influence healing, although molars showed a trend toward lower success at 2 and 3 years. These findings confirm that biological and procedural factors outweigh demographic variables in determining prognosis [74].

While the type of tooth and arch did not significantly predict outcome, restoration type did. During the 5-year follow-up, coronal restoration was significant only at the 1-year recall, at which crowns achieved higher strict success than composites (84.8% vs. 76.7%, p = 0.011). Logistic regression confirmed that crowns were associated with a lower odds of success than composites (OR = 0.59, p < 0.05). At 2 years, crowns showed slightly higher success (88.4% vs. 83.7%) without showing significance (p= 0.090; OR = 1.50, p > 0.05). However, this association disappeared in subsequent recalls, suggesting that initial differences may reflect short-term biological or mechanical factors rather than long-term prognosis. This could be attributed to the fact that crowned teeth are often more structurally compromised and may represent more complex cases, although further investigation is warranted.

Although crowned teeth showed higher strict success rates at the 1-year bivariate analysis, multivariate regression revealed lower odds of success. This apparent discrepancy is likely attributable to case complexity, as teeth restored with crowns may have been more structurally compromised, necessitating indirect restorations. After adjustment for baseline disease severity and other clinical factors, restoration type was no longersignificantly associated with long-term outcomes.

The choice of post-endodontic restoration was based on a clinical evaluation of the remaining tooth structure, including the number of residual marginal ridges and cusp thickness, in line with current evidence-based recommendations [75, 76]. Although coronal restorations were categorized as full-coverage crowns or direct restorations, further stratification among restorative subtypes, such as partial-coverage restorations, was not performed, and therefore their specific influence on treatment outcome was not evaluated.

The number of visits (single vs. multiple) was not associated with the success of treatment, reinforcing the idea that when protocols are standardized and disinfection is properly achieved, procedural variables may have limited influence.

Furthermore, the technical quality of the treatment or retreatment, including factors such as working length control and obturation quality, was not specifically evaluated in the present study. However, all procedures were performed by a single experienced operator following a standardized protocol, which likely ensured a consistent level of technical quality and reduced variability across cases.

Another point worth mentioning is that the reduction in sample size at longer follow-up intervals is not due to patient attrition, but to the temporal distribution of the treatments. Since the 637 cases were treated over 5-year period, only those performed in the earlier years had reached the 4- and 5-year recalls at the time of data collection. Thus, the declining number of cases over time reflects follow-up maturity rather than drop-out bias. No active recall bias was present, as patients were routinely scheduled for annual maintenance visits including radiographic control. However, the reduced number of cases available at the 5-year follow-up represents a limitation of the study, as it may reduce statistical power and require cautious interpretation of long-term findings. Nonetheless, clinical studies with 5-year follow-up on bioceramic sealers are not currently available in the published literature. The existing clinical data on bioceramic sealers demonstrate favorable short- to mid-term outcomes but lack the long-term validation needed for definitive conclusions [34].

Root canal anatomy was not specifically analyzed as a variable in the present study, although both single-rooted and multi-rooted teeth were included. Anatomical complexity may influence the adaptation of filling materials, particularly when using CHC techniques. Experimental studies have shown that in anatomically complex canal systems, such as oval or irregular canals, this technique may result in a higher incidence of voids and reduced material adaptation compared with thermoplasticized techniques. However, despite these laboratory findings, clinical studies have not demonstrated significant differences in long-term outcomes between obturation techniques, suggesting that the bioactive properties of CSBSs may compensate for these limitations [29, 77].

One key strength of this study lies in its methodological rigor. Unlike most published studies on CSBSs, which are often conducted in academic settings and involve multiple operators of varying experience levels, all treatments in this investigation were performed by a single experienced clinician in a private practice setting, using a standardized technique. This reduces variability and enhances the internal validity of the results. Previous studies have highlighted the impact of such heterogeneity on treatment outcomes and the difficulty of generalizing results to real-world practice settings [38, 40].

The present study has some limitations to be acknowledged. First of all, being retrospective, the study lacks randomization and full control of variables, which reduces the strength of causal inferences compared with prospective studies. Second, the absence of a control group represents a limitation of the present study, as it prevents direct comparison with alternative obturation materials and limits causal inference. Therefore, no conclusions can be drawn regarding the superiority of the tested sealer over other materials. However, this choice was intentional, as the primary aim was to assess the healing capacity of this specific material over a 5-year follow-up and evaluate the prognostic factors that influence its success rather than to perform a comparative trial. According to the available literature, the type of sealer (CSBS versus resin-based) and the obturation technique do not significantly influence overall success rates [59]. Therefore, the inclusion of a control group using alternative sealers was not considered essential for answering the present research question. Third, radiographic evaluation relied solely on periapical radiographs, which are less sensitive than CBCT for detecting small lesions or incomplete healing. Furthermore, all treatments were carried out by one experienced operator, which guarantees consistency but may limit generalizability to broader clinical settings.

In addition, the inclusion of both primary treatments and retreatments, along with the wide variability in clinical conditions, may introduce heterogeneity in case complexity and influence treatment outcomes. Although all procedures were performed using a standardized protocol by a single experienced operator, differences in anatomical and clinical difficulty cannot be completely excluded. In this context, the use of standardized case-difficulty assessment tools, such as those proposed by professional endodontic associations, may improve case classification and enhance comparability across studies. Future investigations should consider incorporating such systems to better stratify clinical complexity and refine the interpretation of outcomes. Ultimately, the clinical complexity associated with retreatment procedures may represent a confounding factor, as these cases are often characterized by the presence of previous filling materials, anatomical alterations, and a higher risk of procedural complications. Therefore, both anatomical and clinical variability should be considered when interpreting the present findings.

Within the limitations of this retrospective cohort study, the use of a CSBS was associated with favorable long-term clinical outcomes. However, no conclusions can be drawn regarding the comparative effectiveness of this material.

Future research should continue to follow this cohort over longer periods to provide a clearer picture of the durability of outcomes. It would also be valuable to conduct prospective randomized controlled trials with comparative sealers and standardized follow-up, in order to better define prognostic factors and validate these findings across different clinical settings.

Conclusions

This 5-year retrospective analysis demonstrated that endodontic treatment using a CSBS (CeraSeal) achieved high and progressively increasing success rates, with strict success rising from 80.2% at 1 year to 87.67% at 5 years. Loose criteria consistently yielded higher values, though their sensitivity to minor radiographic alterations was lower. Among the prognostic factors, higher baseline PAI scores were the most consistent predictors of reduced success, while other variables, such as tooth type, extrusion, or restoration, showed no significant influence. Small perforations negatively affected outcomes at 3 and 4 years under strict criteria. Overall, the findings confirm the long-term effectiveness of CSBS in periapical healing and highlight the importance of baseline lesion severity in predicting prognosis.

Author contributions

All authors contributed to the study’s conception and design. **F.C.:** Investigation, Methodology, Data curation, Software. **G.M.** : Writing- Reviewing and Editing, Conceptualization, Supervision **S.S.** : Conceptualization, Methodology, Data curation, Investigation. **M.C.** *:* Conceptualization, Investigation, Writing-Reviewing and Editing, **D.P.:** Writing- Reviewing and Editing, Conceptualization, Supervision. **S.G.** : Writing-Reviewing and Editing, Data curation **C.G.** : Investigation, Supervision, Conceptualization, Investigation, Reviewing and Editing, Supervision.

Funding

No funding was obtained for this study.

Data availability

The data supporting this study’s findings are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The University of Siena Ethics committee approved this study (protocol number: 18202/2020).

Competing interests

The authors declare no competing interests.