Endocrowns vs.Ceramic Onlays Used in Endodontically Treated Teeth- Randomized Clinical Study

Study design - ethical and methodological aspects This study is an integral part of an umbrella project, in which the primary outcome is the longevity of prosthetic treatments, whereas patient-centered secondary outcomes are dealt with in the current study. The study protocol was approved by the Research Ethics Committee, following the Consolidated Standards of Reporting of Trials (CONSORT and CONSORT PRO) guidelines (Calvert et al., 2013; Moher et al., 2012), and registered at clinicaltrials.gov (NCT03064516).

This double-blind, parallel-arm randomized clinical trial, with a 1:1 allocation ratio, was conducted by a previously trained operator, who performed all the restorative procedures.

Survival rate of inlay/onlay ceramic restorations (96.6% after 24 months) was the primary outcome considered for sample size calculation (Peumans et al., 2010). Therefore, a clinically significant difference of 15% is expected for longevity between the groups. So, adopting a significance level of 0.05 and a 0.80 power, considering one tooth per patient, and using a two-tailed test for noninferiority trials, adding 20% for possible losses (dropout), the final number of teeth per group was 20, corresponding to a total of 40 teeth.

Participants - recruitment, eligibility, randomization, and allocation The patients were recruited by one of the researchers (SM) based on history-taking, clinical examination, and radiographic exams using a sample of patients from the Military Police Dental Center (São Paulo, Brazil) treated in 2017. The clinical steps were carried out in a private dental office.

Eligible patients were randomly assigned to one of the parallel arms of the study. The randomization sequence was generated (www.sealedenvelope.com) and allocation concealment was ensured by the use of sealed brown envelopes numbered serially and kept in possession of an independent researcher until they were opened by the operator at the beginning of treatment. The contents of the envelope indicated the experimental group to which the patient would be assigned: Endocrown group (test) or Partial coverage ceramic restoration + fiber post group (PCCR+post- control).

Restorative techniques for the experimental groups All participants received instructions on oral hygiene and diet. Radiographs were taken at baseline and after cementation. Photographs were taken at baseline, after removal of the carious tissue and of old restorations, after cavity preparation, and after cementation. Procedure time was recorded in sessions 1 and 2, beginning with preoperative mouth rinsing and ending with occlusal adjustment (session 1) and beginning with preoperative mouth rinsing and ending with final radiograph (session 2).

Session 1- Cavity preparation and casting Endocrown group- Well-defined cervical chamfered finish lines should be created to facilitate the impression and technical procedures, thus a 2-mm round-ended chamfer finish line was created along the margin using tapered inverted cone diamond burs, at high rotation and under cooling, against a 1:5 contra angle multiplier (S-MAX M95L- NSK). No filling was used, the space of the pulp chamber is included in the preparation.

PCCR+Post group- A fiber post (White post, FGM) whose size was compatible with the main canal was luted, reaching approximately half of the canal. The post was cleaned with alcohol, dried, and treated with Monobond (Ivoclar Vivadent); Multilink was applied to the tooth (Ivoclar Vivadent), mixing it with primer A+B at a 1:1 ratio, without previous acid etching, with a brief air jet. After that, the post was luted with Multilink resin cement (Ivoclar Vivadent), followed by photopolymerization for 20 seconds and filling with composite resin - shade A2 (Tetric N Ceram bulk fill- Ivoclar Vivadent), in 1-2 mm increments and photopolymerization for 20 seconds per layer. Occlusal and proximal boxes were prepared, not exceeding 2 mm in ceramic thickness. A 2-mm round-ended chamfer finish line was created along the margin of extracoronal areas using tapered inverted cone diamond burs, at high rotation and under cooling, against a 1:5 contra angle multiplier (1:5 S-Max M95L, NSK).

In both groups, the cusps were preserved, whenever possible, but in cases where veneering was needed, the occlusal surface was abraded to allow for a 1.5-to-2 mm space. If necessary, the proximal contact point was removed with metal files.

The double-cord technique (Ultrapack- Ultradent) and double molding technique were used (Virtual heavy and regular body - Ivoclar Vivadent). The color was chosen using the Vita classical shade guide (Vita Zahnfabrik). The provisional restorations were fabricated with acrylic resin and cemented with Temp Bond NE (Kerr Corporation).

Laboratory phase - ceramic restoration manufacture The specimens were fabricated with lithium disilicate glass ceramic (IPS e.max CAD- Ivoclar Vivadent) and milled (Cerec In Lab Mcxl), producing a monolithic restoration, which was tinted and glazed.

Session 2- Testing, fitting, and cementation of the ceramic restoration After testing and fitting of the specimen in the mouth, rubber dam isolation was used for adhesive cementation in both groups, as follows: the enamel and dentin were etched with 37% phosphoric acid gel (N-etch; Ivoclar Vivadent) for 15-20 seconds, followed by copious rinsing and brief air jets. The specimens were etched with 5% hydrofluoric acid (IPS Ceramic Etching Gel; Ivoclar Vivadent), silanized for 1 min (Monobond N; Ivoclar Vivadent) and luted with Tetric N-bond and Multilink N dual cement (Ivoclar Vivadent). Photoactivation was performed with a Radii-Cal device (SDI- 1200 mW/cm2) for 20 seconds on each surface. Occlusal adjustment was performed and the restorations were polished with rubber points (Optra Fine Ivoclar Vivadent) at low speed, under cooling.

Patient-centered outcomes

Endodontic treatments had been carried out by different dentists on dates that could not be clearly specified by the patients. Patient-reported outcome measures (PROMs) include self-administered questionnaires, with the purpose of evaluating psychometric data, in addition to history taking and demographic and clinical data. The questionnaires described next were handed out to participants by an external examiner, in the absence of the operator:

• Patient satisfaction questionnaire (Mattos-Silveira et al., 2015) - The patient was asked to give his/her real opinion about the treatment and to indicate how satisfied he/she was on a 0-3 scale: 0 - excellent; 1- good; 2- acceptable; 3- not satisfied.
• Visual analog scale- Used to assess patient discomfort/pain immediately after the end of sessions 1 and 2. This scale consists of a 10-cm (100-mm) horizontal line, with "no sensitivity" written on one end and "maximum sensitivity" on the other one. The patient was asked to draw a vertical line on the horizontal scale to indicate his/her level of discomfort/pain. After that, the distance (in mm) from the starting point (no sensitivity) to the vertical line drawn by the patient was measured with a ruler (0 to 100 mm). The recorded values were interpreted as follows: 0-4 mm no discomfort; 5-44 mm mild discomfort; 45-74 mm moderate discomfort; 75-100 maximum discomfort.

Statistical analysis The data were assessed by SPSS V16 for Windows (SPSS; Chicago, IL, USA). Poisson regression analysis was used to compare patient discomfort between the groups and to evaluate the influence of explanatory variables on the discomfort reported by patients (age and sex, tooth, type of restorative treatment, number of treated surfaces, presence of endodontic lesion, and procedure time). At first, an unadjusted Poisson regression analysis was performed for each explanatory variable, and values with p<0.20 were included in the adjusted regression model. Only those variables with p=0.05 were kept in the final model. Prevalence ratios (PR) were calculated using a 95% confidence interval (95%CI). As the level of satisfaction was 100% for score 0, it was not possible to conduct a statistical analysis. Two-way repeated-measures analysis of variance - experimental group and time - was used to estimate the difference in procedure time between the experimental groups. Kaplan Meyer will be performed to survival analysis.

In all analyses, the significance level was set at 5%.