Predictors of periapical bone healing associated with teeth having large periapical lesions following nonsurgical root canal treatment or retreatment: A cone beam computed tomography‐based retrospective study

Int Endod J. 2024;57:23–36. | 23 wileyonlinelibrary.com/journal/iej Received: 26 June 2023 | Accepted: 18 October 2023 DOI: 10.1111/iej.13993 ORIGINAL ARTICLE Predictors of periapical bone healing associated with teeth having large periapical lesions following nonsurgical root canal treatment or retreatment: A cone beam computed tomography-based retrospective study CarolinaMosquera-Barreiro1 | ManuelRuíz-Piñón1 | Francesc AbellaSans2 | VenkateshbabuNagendrababu3 | Thilla SekarVinothkumar4 | JeniferMartín-González5 | BenjamínMartín-Biedma6 | PabloCastelo-Baz6 1Surgery and medical surgical specialties. International theoreticalpractical Master of Endodontics, Dentistry Restorative and Aesthetics, University of Santiago de Compostela, Santiago de Compostela, Spain 2Department of Endodontics, Universitat Internacional de Catalunya, Barcelona, Spain 3Department of Preventive and Restorative Dentistry, University of Sharjah, College of Dental Medicine, Sharjah, United Arab Emirates 4Department of Restorative Dental Sciences, Division of Operative Dentistry, College of Dentistry, Jazan University, Jazan, Saudi Arabia 5Endodontic Section, Department of Stomatology, School of Dentistry, University of Sevilla, Sevilla, Spain 6Oral Sciences Research Group, Endodontics and Restorative Dentistry Unit, School of Medicine and Dentistry, University of Santiago de Compostela, Health Research Institute of Santiago de Compostela (IDIS), Santiago de Compostela, Spain Correspondence Benjamín Martín-Biedma, Oral Sciences Research Group, Endodontics and Restorative Dentistry Unit, Abstract Aim: Several factors influence the condition of the periapical tissues associated with root filled teeth. The primary objective of this study was to retrospectively evaluate the extent and speed of bone healing of large periapical lesions associated with nonsurgical root canal treatment or retreatment. The secondary objective was to analyse the relationship between the time to complete healing when analysed using cone beam computed tomography (CBCT) and other possible predictors that affect healing. Methodology: Seventy-nine patients were treated during the years 2013–2020 with large periapical lesions of endodontic origin (10–15 mm) as observed on intraoral periapical radiographs (IOPAR) were included. IOPAR and CBCT were available before treatment and during the follow-up (IOPAR every 6 months and CBCT every 12 months). The volume of periapical lesions was calculated by OsiriX Lite software. Variables such as initial volume of the lesion, age, gender, type of treatment or type of root canal filling were compared to identify the differences between healed and unhealed lesions. Pearson's Chi-square test was used for categorical variables, the t-test for age and the Wilcoxon test for initial volume of the lesion. The association between time to healing and the variables was assessed using univariate analysis and multivariate analysis. The Wilcoxon test was used to observe the association of healing time with categorical variables and the correlation index was measured with the quantitative variables. Results: Of the 79 cases analysed, 60 lesions (76%) were completely healed as verified by CBCT in a mean healing time of 19 months, of which 60% healed fully between 12 and 18 months. Increase in age of patient and larger initial volume of the lesion were associated with a significantly longer healing time (p < .001). Gender, filling material and type of treatment did not have a significant effect on the healing process (p > .05). This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2023 The Authors. International Endodontic Journal published by John Wiley & Sons Ltd on behalf of British Endodontic Society. 24 | PREDICTORS ASSOCIATED WITH HEALING OF LARGE PERIAPICAL LESIONS INTRODUCTION Apical periodontitis (AP) is a chronic inflammatory condition that affects the periapical tissues surrounding teeth with infected root canal systems (Nair,2006). Even though AP can be caused by a wide range of etiological factors, it is generally accepted that microorganisms within the root canal system are the primary cause of pulp necrosis and the consequent inflammatory response in the periapical region (Nair,2006). Radiographically, AP is characterized as a radiolucency around the roots of an affected tooth due to damaged periradicular tissues (Jakovljevic etal.,2021). AP is usually diagnosed during routine radiographic examination since it is often asymptomatic (Kruse etal.,2014). Hence, radiographic assessment is crucial in terms of both diagnosis and follow-up of the healing process after treatment (Bergenholtz etal.,2010). Periapical lesions are routinely monitored using intraoral periapical radiographs (IOPAR) (Lo Giudice etal.,2018; Metska et al., 2013; Ørstavik & Pitt Ford, 2008). However, periapical and panoramic radiographs have inherent limitations, including anatomical noise (Bender & Seltzer,2003) and some degree of geometric distortion (Forsberg & Halse,1994), which may hinder the accurate detection of periapical lesions within cancellous bone (Abella etal.,2014; Davies etal.,2016), particularly when the buccal cortical bone is thick (Lo Giudice etal.,2018). These limitations can largely be overcome through cone beam computed tomography (CBCT), which, despite providing a lower spatial resolution than periapical radiographs, provides a three-dimensional (3D) visualization of structures (Kruse etal.,2014). Management of AP is determined by the presence of periapical pathosis and its change over time (Ng etal.,2011). An increase in the size of lesion after root canal treatment or retreatment indicates that other treatments such as apical surgery or intentional replantation are necessary to save the tooth. In contrast, a decrease in the size of the lesion or its complete disappearance is synonymous with healing (European Society of Endodontology,2006). Periapical healing is an essential process to restore tissue integrity (Holland etal.,2017), particularly in large periapical lesions that may cause cortical expansion, cortical plate erosion or inflammatory root resorption of the surrounding teeth (Sharma et al., 2018). It is worth mentioning that a range of therapeutic factors (e.g. type and quantity irrigant solution, type of intracanal dressing or root canal filling), patient's systemic condition and physiology (e.g. chronic diseases or age) can interfere with the periapical healing process and affect the prognosis of root canal treatment (Holland etal.,2017; Georgiou etal., 2020; Jakovljevic etal., 2020). Clinicians do not expect all large periapical lesions to heal completely after root canal treatment, particularly when there are evident borders radiographically mimicking periapical bone cysts (Saini etal.,2023). However, it is not possible to differentiate clinically or radiographically whether the periapical lesion is a granuloma or a cyst (White etal., 1994). This information can only be acquired from an accurate histopathological diagnosis of the entire lesion and root apex (Ricucci etal.,2020). Although histological examination remains the gold standard for diagnosing the presence and severity of AP, the invasiveness of the procedure impedes its routine use (Kruse etal.,2014). Consequently, diagnosis of AP is carried out clinically and radiographically. Evaluation of a periapical lesion is recommended until it has healed or for at least 4 years post-treatment (European Society of Endodontology, 2006). Persistence of a lesion after 4 years indicates that root canal treatment is associated with a post-treatment disease, when additional treatment should be considered (European Society of Endodontology,2006). The main aim of the present retrospective study was to evaluate the healing of large periapical lesions and determine the minimum follow-up required after a nonsurgical root canal treatment or a nonsurgical retreatment. A secondary objective was to analyse the effects of initial volume of the lesion, age and gender of the patient, type of treatment (root canal treatment or retreatment) or type of root canal filling (gutta-percha or calcium silicate-based material) on the healing time. School of Medicine and Dentistry, University of Santiago de Compostela, Health Research Institute of Santiago de Compostela (IDIS), Santiago de Compostela 15706, Spain. Email: [email protected] Conclusions: Clinicians should be aware that periapical lesions in older patients and larger areas of bone loss take longer to heal. CBCT monitoring of large periapical lesions is critical and it can help clinicians in the decision-making process. KEYWORDS cone beam computed tomography, follow-up, periapical healing, root canal treatment 13652591, 2024, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/iej.13993 by Universidade de Santiago de Compostela, Wiley Online Library on [22/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | 25 MOSQUERA-BARREIRO etal. MATERIALS AND METHODS Ethics and reporting guidelines The current study design was approved by the Research Ethics Committee (Santiago, de Compostela – Lugo (Galicia, Spain), approval No: 2021/508). The current study was reported according to Preferred Reporting items for Observational studies in Endodontics (PROBE) 2023 guidelines (Nagendrababu etal.,2023). Sample size estimation A random sample of 79 individuals was sufficient to estimate, with a confidence level of 95% and a precision of ±10.7 percentage units, a population percentage that was predicted to be around 63%. An estimate of a 63% healing rate was made based on 24 months of data (Zhang et al., 2015). The calculations were made using the GRANMO sample size calculator. (https:// www. imim. es/ ofert adese rveis/ softw arepublic/ granmo/ ). Case selection criteria In this retrospective study, images of IOPAR/CBCT and clinical data belonged to 79 patients from the Masters of Endodontics, Restorative and Aesthetic Dentistry (University of Santiago de Compostela, Spain) in collaboration with five private clinics (Endodontic practice, Spain). Participants were informed of the purpose of the study and provided written consent. Seventy-nine patients were selected by five operators who participated in this study. The inclusion criteria were that patients had an initial periapical lesion on IOPARs ranging in size from 10 to 15 mm at their greatest diameter (Calişkan,2004; Saini etal.,2023). Root canal treatment or retreatment was performed between January 2013 and December 2020 with radiographic follow-ups every 6 months and CBCT every 1 year until healing or when further treatment was deemed necessary up to 48 months. The exclusion criteria were teeth with limited remaining tooth structure, cases of endodontic microsurgery, teeth extracted for nonendodontic reasons and teeth with signs or symptoms of post-treatment disease during the follow-up period of 48 months. Additionally, teeth were excluded if they had previous periodontal disease, underwent prior surgical endodontic treatment, or if the apex was not visible on any of the radiographs. Teeth were excluded from the analysis of ‘periapical status following treatment’ if details on the periapical status at the time of extraction were unavailable and an adequate pre and intraoperative dataset was not obtained for each tooth. Preoperative radiological assessment and volumetric measurement All radiographs were takes using ‘as low as diagnostically acceptable’ (ALADA) principle and standard imaging protocol. The IOPAR were exposed with an X-ray generator (VistaIntra DC, Durr Dental Médics Ibérica, Barcelona, España) and digital imaging plates (Soredex, Tuusula, Finland) under the following parameters: 65 kV, 7 mA and 0.12 s exposure time. IOPAR was taken with the paralleling technique using a film holder (Rinn XCP Ring® positioner, Dentsply, Constanz, Germany). The exposed imaging plates were introduced into the scanner (CS 7600; Carestream Dental LLC, Atlanta, Georgia) and were read using the CS Imaging version 8 software (Carestream Dental LLC). The IOPAR of each patient was assessed on a 23.8-inch flat screen monitor (HP Pavilion, Palo Alto, CA, USA) in a quiet, dimly lit room and the greatest diameter of periapical lesion was recorded in. Initial 2D measurements were made by two endodontists, the kappa value for the initial agreement was 0.75; cases in doubt were resolved by discussion. Adjustments in contrast or magnification of images during the observation were avoided to prevent subjective changes in the dimensions of lesion as they were captured under standard radiographic exposure conditions. Subsequently, a CBCT scan was performed with the Carestream CS9300 Select equipment (CS 9300, Carestream Dental LLC, Atlanta, Georgia, USA) following the manufacturer's instructions under standard exposure settings (84 kV and 5 mA) and patient position. The voxel size was 0.18 mm, with a field of view (FOV) of 5 × 5 cm and an exposure time of 20 s. The images were reconstructed with the CS 3D Imaging Software (Carestream Dental LLC) and exported using the DICOM format to a 3D planning software to perform the volumetric measurements (OsiriX Lite software, Pixmeo, Geneva, Switzerland; www. osirix viewer. com). Volumetric measurements were made in the axial plane to visualize the lesion in its maximum extension in the apicocoronal direction. A ‘closed polygon’ tool was used to delimit the border of the lesion whilst scrolling the slices from the beginning of the lesion (Figure1a). The border of the lesion was demarcated in each slice (Figure1b) to calculate the volume (Figure1c). The volumetric measurements of the periapical lesions before treatment and during follow-ups were recorded on an Excel spreadsheet (Microsoft Excel 2016, Microsoft Corporation). Assessment of medical records The following patient-specific information was recorded: name, file number, the tooth involved, age, gender, type of 13652591, 2024, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/iej.13993 by Universidade de Santiago de Compostela, Wiley Online Library on [22/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 26 | PREDICTORS ASSOCIATED WITH HEALING OF LARGE PERIAPICAL LESIONS treatment (root canal treatment or retreatment) and filling material (gutta-percha or calcium silicate-based material). Clinical procedures The root canal treatment and retreatments were performed by five endodontists with more than 8 years of experience, who collected the preoperative, intraoperative and postoperative information following the same protocol. The treatments were carried out under rubber dam isolation and using an operative microscope. The access opening was performed with Endo Access burs (Dentsply Tulsa Dental Specialties; Tulsa, Oklahoma, USA) and Endo-Z burs (Dentsply, Maillefer, Ballaigues, Switzerland). In primary endodontic infections, the patency of all root canals was verified with manual files K-Flexofile (Dentsply Maillefer, Ballaigues, Switzerland) K-File (Endogal, Sarria, Lugo, Spain). After an adequate glide-path was established, the rotary instrumentation systems, ProTaper Gold (Dentsply Maillefer, Ballaigues, Switzerland) or Endogal (Sarria, Lugo, Spain) were used. The nonsurgical retreatments were performed with ProTaper retreatment files (Dentsply Maillefer; Ballaigues, Switzerland) and Endogal files. Mechanical instrumentation of canals was carried out with abundant and regular use of sodium hypochlorite at 5.25% (Dentaflux, Madrid, Spain) with ultrasonic activation. The treatment was carried out in two sessions with intracanal medication between visits (Endocalcium, Endogal, Sarria, Lugo, Spain). Finally, the canal system was filled using the continuous wave technique and resin cements, AH Plus sealer (Dentsply DeTrey GmbH, Konstanz, Germany) or Endoresin cement (Endogal, Sarria, Lugo, Spain). The calcium silicate-based materials, ProRoot MTA (Dentsply Maillefer, Ballaigues, Switzerland), Biodentine (Septodont, Saint-Maur-des-Fossés, France) and EVO MTA (Endogal, Sarria, Lugo, Spain), were also used in those canals where the apical constriction had a diameter of greater than 0.5 mm. The temporary coronal restoration between visits was performed with polytetrafluoroethylene (PTFE) tape (Diamond, India, ISO 9001:2000, 12 mm) and Grandioso Flow (GRAFL; VoCo, Cuxhaven, Germany) until the coronal restorations were placed. Clinical evaluation and postoperative radiological evaluation Patients were evaluated periodically at 6-month intervals for the presence or absence of signs and symptoms such as pain on palpation/percussion and swelling. At each follow-up visit, an IOPAR was taken under the same standard conditions and compared with the pre-treatment IOPAR for the greater diameter of the periapical lesion. Similarly, after every 12 months, in addition to the IOPAR, a CBCT was taken to compare the volume of the lesion. A confirmatory CBCT was performed during the specific 6 month's follow-up where no periapical lesion was observed on the IOPAR. The integrity of the final restoration was assessed during each visit. Two blinded endodontists assessed the pre and postoperative IOPAR and CBCT images. By giving each image/ volume a random number and arranging them in a random order, the observers were blinded on the treatment status of patients. To calibrate the examiners prior to the FIGURE 1 (a) Initial position before first volumetric measurement. (b) Delimitation of the lesion in the axial plane for volumetric measurement. (c) Calculation of the initial volume. 13652591, 2024, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/iej.13993 by Universidade de Santiago de Compostela, Wiley Online Library on [22/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | 27 MOSQUERA-BARREIRO etal. study, they examined 10 volumes independently. After evaluation, a kappa value for agreement was calculated (weighted kappa values, k = 0.8), when discrepancies occurred the observations were compared and discussed in order to reach an agreement. The treatment was considered a success when there was no periapical lesion on the CBCT images. Absence of periapical lesion was recorded when the reconstructed volume of the lesion in on the CBCT image was less than twice the width of the periodontal ligament (Low etal.,2008; Metska etal.,2013) and the volume was noted as zero. If the volume of the lesion had decreased or remained the same, follow-up visits were continued for up to 4 years. To determine the progress of healing comparisons were always made with the volume of the lesion calculated on the CBCT image of the previous follow-up. In the fourth year, if a tooth still had a periapical lesion, it was considered a failed treatment. It was also considered a failed treatment when the lesion size increased during the follow-ups and if an extraction was necessary due to fractures or periodontal defects with a poor prognosis. The healing of the lesion was dichotomized as completely healed or not healed based on the above protocol. The postoperative radiological evaluation is described in Figure2. Statistical analysis The statistical analysis was performed using SPSS software (v.20; IBM). The quantitative variables (age, volume and healing time) were described using position statistics (mean and median) and dispersion (range, interquartile range and standard deviation). The qualitative variables (gender, type of filling material and type of treatment) were described using absolute and relative frequencies. The healed and unhealed cases were compared to determine any differential factors between the two groups. Pearson's Chi-square test was used for categorical variables. To choose the tests for comparison of quantitative variables, the normality and homoscedasticity of the variables were evaluated using Anderson–Darling and Fligner-Killen tests, respectively. Student's t-test was used when normality and homoscedasticity assumptions were met, Welch's t-test was used when there was heteroscedasticity and Wilcoxon–Mann–Whitney test was used when normality assumptions were violated. The association of healing time with the rest of the variables included in the data set amongst the healed patients was evaluated through a univariate analysis (variable by variable). The Wilcoxon test was used to observe the association of categorical variables with healing time. In the association of time with quantitative variables, the Pearson correlation coefficient was measured. Nonparametric (i.e. Turnbull's Estimator) and accelerated failure time models were adjusted for healing times obtained from 6-months intervals and using the icenReg package (Anderson-Bergman, 2017) of R 4.0 (R Core Team, 2021). Total healing time is defined as the interval between visits during the patient's recovery. Different distributions were tested for time until the one that the best fit with the data according to the Akaike Information Criterion (AIC) was found. For the selection of variables that were included in the model, the stepwise methodology of selection of covariates according to the AIC was used and all those that were not significant were eliminated from the model. The survival model coefficients were associated with the healing time with an exponential relationship. The level of significance was established at p < .05. RESULTS Of the 180 screened patients, data from 79 patients were used of which 41 (52%) were men and 38 (48%) were women, whose mean age was 44 (SD = 16.3) years old. In relation to the type of treatment, 47 (60%) cases were retreatment and 32 (40%) were primary root canal treatment. The process of inclusion and exclusion of cases for the purpose of participation eligibility in the current study and the case follow-up is described in Figure3. Overall, 60 patients had complete healing of the periapical lesion with a mean healing time of 19 months (Figure4) Complete healing of the periapical lesion did not occur in 19 patients (Figure5); 14 of them underwent periapical surgery of the affected teeth because of the increase in lesion size during the follow-up period and five of them underwent extraction (two due to traumatic fractures and three due to periodontal defects). The clinical data and the volumetric measurements of each periapical lesion during its follow-up are presented in Table1. The main statistically significant difference between patients with complete healing of the periapical lesion and those without was age (Student's test t 77 = −2.81, p < .01). Patients with complete healing of the periapical lesion had a mean age of 41.2 years (SD = 15.3) and unhealed ones of 52.8 years (SD = 16.7). Furthermore, a first correlation analysis revealed an increase in healing time according to the initial volume of the periapical lesion in mm3 (rPearson = 0.40, p < .01). Therefore, increased age and larger initial volume of the lesion were significant predictors for the healing of periapical lesions. Subsequently, using a parametric survival model, the effect of the covariates on healing time was evaluated. The distribution chosen for the time was gamma by offering a better fit and minimizing the AIC, compared to the 13652591, 2024, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/iej.13993 by Universidade de Santiago de Compostela, Wiley Online Library on [22/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 28 | PREDICTORS ASSOCIATED WITH HEALING OF LARGE PERIAPICAL LESIONS exponential one (Figure6). More than 60% of the patients displayed healing between months 12 and 18. Therefore, in this analysis, it was confirmed that older age and larger initial lesion size was associated with longer healing time (p < .001). Specifically, given the regression coefficients for age (βAge = 0.048, e0.048 = 1.049) and initial volume of the lesion (βVol = 0.003, e0.003 = 1.003), it is concluded that a 1-year increase in age increased the healing time by 4.9% and that an increase in 1 mm3 in the initial volume of the lesion increased healing time by 0.3% whilst the other FIGURE 2 Flow chart of postoperative radiological evaluation. 13652591, 2024, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/iej.13993 by Universidade de Santiago de Compostela, Wiley Online Library on [22/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | 29 MOSQUERA-BARREIRO etal. variables remained constant (Table2). The marginal effect of age and lesion size on healing time at its mean value can be seen graphically in Figures7 and 8. The final model with two covariates minimizes the AIC and showed an acceptable predictive performance (R2 = 0.43). Gender, type of filling material, and type of treatment had FIGURE 3 Flow chart of the study outline. FIGURE 4 Multiplanar reconstructed CBCT images of a healed case at 24 months follow-up. Preoperative (a–d), 12 months follow-up (e–h) and 24 months follow-up (i–l). 13652591, 2024, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/iej.13993 by Universidade de Santiago de Compostela, Wiley Online Library on [22/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 30 | PREDICTORS ASSOCIATED WITH HEALING OF LARGE PERIAPICAL LESIONS no effect on healing (p > .05) in the complete model and were excluded from the final model after applying a variable selection step. DISCUSSION In the current study, the post-treatment evaluation of 79 patients with a periapical lesion ranging in size from 10 to 15 mm at their greatest diameter (Calişkan,2004; Saini et al., 2023) was evaluated retrospectively. Monitoring each periapical lesion with CBCT imaging allowed volumetric measurements to have an objective and quantifiable view of the process. The main objective to follow-up a periapical lesion after primary root canal treatment or nonsurgical retreatment is to observe the healing process and rule out its persistence (European Society of Endodontology,2006). The aim of this study was to evaluate the successful healing of large periapical lesions and to determine the minimum required follow-up time. Furthermore, the purpose was also to analyse the possible factors that could influence this healing process. FIGURE 5 Multiplanar reconstructed CBCT images of an unhealed case at 12 months follow-up. Preoperative (a–d), 12 months followup (e–h). TABLE 1 Descriptive analysis between healed and not healed patients, and total. Variable Healed (N = 60) Not healed (N = 19) Total (N = 79) p-Value Sex N60 19 79 .55 (a) Male 30 (50.0%) 11 (57.9%) 41 (51.9%) Female 30 (50.0%) 8 (42.1%) 38 (48.1%) Age (years) N60 19 79 <.01 (b) Mean (SD) 41.23 (15.29) 52.79 (16.68) 44.01 (16.30) Median (IQR) 38 (25.5) 56 (24.5) 40 (24.5) Treatment N60 19 79 .36 (a) RCT 26 (43.3%) 6 (31.6%) 32 (40.5%) NSR 34 (56.7%) 13 (68.4%) 47 (59.5%) Material N60 19 79 .75 (a) Calcium silicate-based material 8 (13.3%) 2 (10.5%) 10 (12.7%) Gutta-percha 52 (86.7%) 17 (89.5%) 69 (87.3%) Initial volume (mm3) N60 19 79 .76 (c) Mean (SD) 222.49 (225.75) 215.53 (276.43) 220.81 (237.05) Median (IQR) 113 (270.4) 102.9 (130.65) 108.3 (226.4) Note: Tests used: (a) chi-square; (b) Student's t-test; (c) Wilcoxon–Mann–Whitney test. SD (standard deviation). 13652591, 2024, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/iej.13993 by Universidade de Santiago de Compostela, Wiley Online Library on [22/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | 31 MOSQUERA-BARREIRO etal. A standard protocol was followed for all cases by both the faculty of university clinics and endodontist of private clinics, where a medication with calcium hydroxide (Best etal.,2021) is placed for a month, to check resolution of sinus tract, fistulous process, or the abscess (if there is one) and the disappearance of pain. These factors are favourable clinical signs of effective root canal disinfection (Ordinola-Zapata etal.,2022). The outcome of the treatment was that 60% healed between 12 and 18 months. The study by Metska etal.(2013) revealed, after evaluating the volumetric changes, a significant reduction in the size of periapical lesions 1 year after root canal retreatment in more than 50% of the teeth. Zhang etal.(2015) observed that the rate of reduction of periapical lesions after endodontic treatment was higher during the first-year of follow-up. They stated that the greatest decrease occurred in the time immediately after the treatment was performed (Zhang etal.,2015). Weissman et al., 2021 evaluated 384 teeth postoperatively for a follow-up period of 48 months using 2D radiographic images and concluded that the healing process of periapical lesions at 12 months provide a good indication of the long-term result of the treatment. However, follow-up with 2D radiography alone, as is well known, has inherent limitations, including anatomical noise (Bender & Seltzer,2003) and some degree of geometric distortion (Forsberg & Halse,1994), which may hinder the accurate detection of periapical lesions within cancellous bone (Abella et al., 2014; Davies et al., 2016), particularly, when the buccal cortical bone is thick (Lo Giudice etal.,2018). These limitations may mask an increase in the size of the lesion and delay further appropriate treatment. Objective monitoring of these lesions on the basis of low radiation is also crucial as over-radiation could compromise the overall health of the patient. Healing of a periapical lesion of endodontic origin is a dynamic process and requires sufficient time to assess its progression and completion (Zhang etal.,2015). The absence of pain, inflammation, loss of function and radiographic evidence of normal periodontal tissues indicate a favourable treatment outcome. The root canal treatment is considered uncertain when radiographically the lesion has remained the same size or has decreased in size, but not completely disappeared. If a lesion persists after 4 years, the root canal treatment is usually considered to be associated with post-treatment disease (European Society of Endodontology,2006). Exceptionally, an extensive radiolucent lesion may heal but leave an irregularly shaped demineralized area visible locally. This defect may be the formation of scar tissue rather than a sign of persistent apical periodontitis and the tooth should be further evaluated (European Society of Endodontology,2006). Furthermore, differential diagnosis of radiographically identified periapical lesion is important to consider as not every periapical lesion is a dental granuloma that will heal. Bhaskar(1966) studied 2308 cases with radiolucency in the apical areas of the teeth. Nine different types FIGURE 6 Baseline survival graph of time to complete healing of the lesion (S(t)) over time (months). Continuous lines represent the Turnbull's nonparametric estimator for the survival (SP) and dashed line represents the parametric model with a Gamma distribution (G). Coefficient Exponential (Coefficient)aSE z-Value p-Valueb Age (years) 0.05 1.05 0.01 3.19 .002** Initial volume (mm3) 0.003 1 <0.01 3.25 .001** Log (shape) 0.65 1.92 0.24 2.77 .006** Log (scale) 2.26 9.59 0.26 8.63 <.001*** Note: ** = p ≤ .01; *** = p < .001. aExponential of the regression coefficient. bSignificance. TABLE 2 Coefficients of the adjusted parametric survival model after the variable selection step, with its standard error, the z-test statistic and the associated p-value. 13652591, 2024, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/iej.13993 by Universidade de Santiago de Compostela, Wiley Online Library on [22/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License