Survival and Success of Two Different Filling Techniques

The study was designed in january 2023 as a prospective randomized clinical study. No major modifications were made to the study design after its initial conception. The patients were treated in the Endodontic Clinical Section-Dental Clinic, University of Bologna, by a pool of postgraduate master operators in accordance with standardized protocols and under the strict supervision of the experienced tutors of the master. All the operators, before the study started, were adequately instructed and trained in sealer application and obturation technique. The study was approved by the ethical committee (597-2022-SPER-AUSLBO).

The study adhered to the principles of the Declaration of Helsinki, as modified in 2013. The clinical staff provided written and verbal information to patients before enrolment.

All patients provided a signed informed consent to accept the treatment plan and to follow the hygiene program. The study was designed in compliance with the STROBE checklist and the guidelines published by Dodson in 2007.

Primary Root Canal Treatment Nerve block anaesthesia (1.7 mL, mepivacaine chloridrate, Scandonest 3%, Septodont, St.-Maur-des-Fosses, France) and local anaesthesia (1.8 mL mepivacaine chloridrate, Scandonest 2% with 1:100,000 adrenaline, Septodont, St.-Maur-des-Fosses, France) were performed. The total duration of each endodontic session was 60 to 90 min. Dental dam isolation was positioned on the affected tooth. A straight-line access was performed with a diamond bur mounted on high-speed water-cooled handpieces (Cefla, Imola, Italy). A preoperative working length was estimated using periapical radiographs. The crown-down technique was used. Gates-Glidden burs #2 and #3 were utilised when necessary, only in the coronal third. NiTi instruments were used to shape the canals in the coronal, medium, and apical third (Rotate, VDW, Munchen, Germany). An electronic apex locator (Root ZX, Morita, Osaka, Japan) with K-file #10 was used to determine the working length during the entire clinical procedure. Intra-oral periapical X-rays were performed to confirm the working length during the root canal instrumentation. Each root canal was shaped in the apical third with an apical diameter of #25.04 at least.

Irrigation was performed after the use of each instrument with a total of 5 mL of 5% NaOCl solution (Niclor 5, OGNA, Muggiò, Italy).

Secondary Root Canal Treatment An initial pathway was created with Gates-Glidden burs #3 and #4 (Dentsply Maillefer, Ballaigues, Switzerland) to approximately 3-4 mm depth in the gutta-percha. Reciprocating NiTi instruments (Reciproc Blue, VDW, Munchen, Germany) were then used with Silver Reciproc Endomotor in the "Reciproc All" setting. After each step, the material entrapped among the instrument threads was removed using a sterile sponge. The working length was established after the removal of root canal remnants using periapical X-ray and an electronic apex locator. An apical enlargement was performed with Reciproc Blue #40 and #50 when needed. Irrigation was performed using a total amount of 5.0 mL of 5% NaOCl. When necessary, a dental surgery microscope (OMS3200 Dental Microscope, Zumax Medical Co., Suzhou, China) was used to detect the access to root canal orifices and to identify the presence of remnants.

Root Canal Filling Procedures The randomization was performed before the root canal obturation techniques (1:1 ratio). The operator did not know which sealer was used before that time.

Test Group_ A premixed CaSi-containing bioceramic sealer (Ah Plus Bioceramic, Dentsply, Konstanz, Germany) was used in association with a warm carrier-based technique (Thermafil, Dentsply, Konstanz, Germany). AH Plus Bioceramic is mostly composed of zirconium dioxide (50-70%) as a radiopacifier and tricalcium silicate (10-15%) as a bioactive component. Dimethyl sulfoxide and traces of lithium carbonate and thickening agents are also reported by the manufacturer.

The sealer was applied with a sterile K-file inserted into the canal to reach the WL-3 mm and gently moved around the root canal walls. The carrier was heated using a dedicated obturation oven (Thermaprep obturation, Dentsply, Konstanz, Germany) and slowly inserted into the canal at WL-0.5 mm. The excess of the carrier was cut with a round bur.

Control group_ Alternatively, the same protocol was applied with an epoxy resin based sealer.

An X-ray was performed to verify the quality of the root canal obturation. Finally, a small cotton pellet and a temporary restoration (Coltosol, Coltene, Altstaetten, Switzerland) were positioned in the access cavity and maintained until definitive restoration. In case of severe pain, a medical prescription to take NSAID medications (such as ibuprofen or ketoprofen) was prepared by the university staff. In this case, the event was recorded, and the patient was excluded from the study.

Tooth Restoration Teeth were definitely restored within 2 weeks under rubber dam isolation. Temporary restoration was removed using ultrasonic tips, and a crown was restored under rubber dam isolation. Self-etching dentinal bonding agent primer and bonding (Clearfil SE BOND, Kuraray, Osaka, Japan) were applied, photo-cured (Elipar, 3M ESPE, St. Paul, MN, USA) for 30 s and layered by flowable (G_Aenial Flow, GC Corporation, Tokyo, Japan) and composite (G-Aenial, GC Corporation, Tokyo, Japan) resins applied incrementally with 1.5 mm layers.

Radiological Evaluation X-rays were taken after the root canal filling using a parallel technique. The following parameters were used: the target-film distance was approx. 30 cm, 0.41 s exposure at 70 Kw and 8 mA. The radiographs were developed in a standard developer unit at 20 °C (Euronda s.p.a., Vicenza, Italy), 12 s developing time, and 25 s fixing time according to the manufacturer instructions.

Intra-oral periapical X-rays and clinical criteria were used to classify the final outcome, with each patient monitored at 1, 6, 12 and 24 months of follow-up. The root canal obturation was considered "adequate" when the filling material was detected at 0-1.0 mm from the radiological apex. Overfilling, short filling, and sealer extrusion were recorded. X-rays were digitalised using a slide scanner with a mean resolution of 1000 dpi and a magnification factor of 20×.

Periapical Index (PAI) was used to score the preoperative diagnosis and endpoint evaluations, which were evaluated in a double-blind manner by two operators (university researchers trained in this analysis) who did not perform the root canal treatment. PAI calibration was performed using well-defined instructions and periapical radiographs with different periapical lesion scores. To ensure their reliability, the evaluators independently assessed the X-rays. In the event of any discrepancies between their assessments, these were extensively discussed until a mutual consensus was achieved. Sealer extrusion was recorded and measured on each periapical X-ray using open-source software (Image J, Bethesda, MD, USA).

Post-Operative Pain Assessment Post-operative pain was assessed as Patient Reported Outcome (PRO) using a 10 cm Visual Analogical Scale, divided into 0-100 steps, with 0 indicating no pain and 100 indicating the most intense pain. Post-operative pain was evaluated after root canal filling (T0), after 1 day (T1), after 7 days (T7), after 1 month (T28), and after 12 months (T365).

When a tooth presented a PAI 1 or 2, the tooth was considered "radiographically healed".

When a tooth presented an improvement in PAI, the tooth was considered as "radiographically healing".

Statistical Methods Variables were summarised as counts and percentages. Barnard CSM (Convexity, Symmetry, and Minimization) test was used to assess the relationship of each potential prognostic factor with PAI at 12-month follow-up (1-2 [healed] vs. ≥3 [still healing]). Barnard test is an exact unconditional test recommended for association in 2 × 2 tables due to its power and preservation of test size. Operationally, it starts with the most extreme table and sequentially adds more extreme ones based on the smallest p-value calculated by iteratively maximising the probability of a 2 × 2 table. Effect sizes were expressed as differences in percentages with 95% confidence intervals (CIs) derived by matching Barnard CSM p-values.

The 95% CIs for healing and survival rates were obtained with the Bayesian-derived Jeffreys method. The significance level was set at 0.05, and all tests were two-sided. Data analysis was performed with the "Exact" R package.