Many studies have shown that endodontic treatments can provide a high rate of success despite the complexity of the endodontic space [
So, for a variety of reasons, endodontic failures still occur and some practitioners delegate endodontic treatments to qualified endodontists. Therefore, some patients are referred to the Graduate Endodontic Clinic at the Bretonneau Hospital (Paris, France) for both initial treatments and retreatments. One of the criteria of the students’ evaluation is the outcome of the endodontic treatments. The management team of the hospital wanted to study the effectiveness of this clinic too; and a retrospective study was carried out. The purpose of this study was to (a) assess the 1–4-year outcome of endodontic treatment performed by postgraduate students and (b) to examine the outcome predictors after a minimum follow-up period of 1 year [
A total of 296 patients (363 teeth) were treated in the Dental Clinic of Bretonneau Hospital between January 2007 and December 2010 by endodontic treatment (initial or retreatment) on one tooth or more. The sample population included all patients referred to us from general dental practice and other clinical units of the dental hospital except those who had a medical contraindication. The detailed medical and dental history was obtained from each patient. Patients were informed about the various treatment alternatives and the benefits and risks associated with each solution. Informed consent was signed by all patients before treatment.
The exclusion criteria were the patient with high risk of bacterial endocarditis or immunocompromised patients. All treated patients were in good general health, except one of them (Hodgkin lymphoma).
The treatments were performed by 9 postgraduate students, supervised by qualified endodontists under operative microscope (Zeiss Pico WIPO). Aseptic techniques were systematically observed, with rubber-dam isolation, and if needed, reconstruction of missing walls with glass ionomer cement (Fuji IX, GC), and possibly a copper ring.
The root canal preparation was carried out in accordance with the biological and mechanical principles of Schilder [
Pre, intra-, and postoperative X-rays were taken with a Rinn film-holder and operated with the VistaScan digital system (Dürr Dental). Sonic and/or ultrasonic inserts (Endo-Success kit on a suitable device PMAX, Satelec-Acteon) were used if needed.
For retreated teeth, silver points and fragments of broken instruments have been bypassed preferably with hand instruments and vibrated with ultrasonics tips. If a bypass has been obtained and if the instrument could not be removed, it was included in the filling material. Gutta-percha and sealer were removed using manual and ultrasonic files. Solvents (A and B Eugesolv, Septodont) were used to facilitate root canal filling removal if necessary. The smear layer was removed by EDTA (Chelasolv, Septodont) and the canals were dried with sterile paper points. Perforations were filled with Mineral Trioxide Aggregate (MTA ProRoot, Dentsply).
For teeth with preoperative apical periodontitis, the treatment was performed in one session if the chemomechanical preparation of the root canal system was complete, if the tooth was asymptomatic before the appointment, and if the canal was dry. If not, calcium hydroxide was introduced into the canals by hand file or using a Lentulo file.
Root canals were filled by vertical compaction using the Shilder technique [
Glass ionomer cement (Fuji IX Fast, GC corp.) was placed as temporary coronal seal and it was recommended to the patient to see his dentist to implement quickly coronal restoration. In some cases, treatment was protected by a temporary crown sealed by polycarboxylate cement (Durelon, 3 M ESPE).
All patients were treated under the same conditions (same dental office, same materials and instrumentation, and same operating protocols). The only variables were the operator and the patient.
Patient listing was made from postoperative written reports. The telephone numbers of patients were registered with the software Agenda Web (AP-HP). Patients were only contacted by telephone; a message was left on answering machine if no response. In case of missed call, we tried two attempts to contact the patient.
Recall was also standardized and made by two calibrated senior practitionners.
Two independent investigators studied all the anonymized X-Rays and were previously calibrated against a set of 100 reference teeth. Afterwards, they examined all the radiographs independently. A third observer, a highly experienced endodontist, was consulted for cases for which disagreement occurred.
Clinical and radiographic pre-intra and immediate postoperative data were normally recorded in the patient medical record of each patient at the time of treatment by the operator (postgraduate practitioners supervised by the Professor director of the postgraduate cursus). Patients were then invited to a clinic/radiographic recall. Data previously recorded and subsequent data recorded at recall were pooled and evaluated.
Demographic characteristics (age and gender) and tooth characteristics (tooth location, type of tooth, and number of roots) were recorded (Figure
Tooth specifications. (a) For the root canal treatment (23 teeth): tooth location (A), tooth type (B), and number of roots (C). (b) For the root canal retreatment (108 teeth): tooth location (A), tooth type (B), and number of roots (C).
Preoperative conditions collected were clinical signs and symptoms, pulp vitality, presence and size of Apical Periodontitis (AP), root filling material, root filling length, root filling density, presence of a perforation and/or fractured instrument, and time elapsed since the initial endodontic treatment.
Intraoperative characteristics recorded were number of treatment sessions, root filling length, root filling density, presence of a perforation, and/or fractured instrument.
Postoperative characteristics were clinical signs and symptoms, presence and size of periapical radiolucency, type of coronal-root restoration, presence of post, restoration leakage, and fracture and were recorded the day of postoperative control. This was carried out between 1 and 4 years after endodontic treatment for each tooth treated. Clinical and radiographic evaluation was performed by two reviewers. The radiographs were scored according to the PAI system [
In multirooted teeth, the condition of the most severely affected root was considered (Table
Distribution of variables: preoperative, intraoperative, postoperative factors.
Initial treatment | Retreatment | |||
---|---|---|---|---|
|
% |
|
% | |
|
||||
Age | ||||
≤45 years | 9 | 47 | 40 | 47 |
>45 years | 10 | 53 | 46 | 53 |
Gender | ||||
Male | 13 | 68 | 51 | 59 |
Female | 6 | 32 | 35 | 41 |
Tooth type | ||||
Anterior | 4 | 17 | 4 | 4 |
Posterior | 19 | 83 | 104 | 96 |
Tooth location | ||||
Maxillary | 14 | 61 | 66 | 61 |
Mandibular | 9 | 39 | 42 | 39 |
No. of roots | ||||
Single root | 7 | 30 | 30 | 28 |
Multi-rooted | 16 | 70 | 78 | 72 |
Signs and symptoms | ||||
Absent | 20 | 87 | 66 | 61 |
Present | 3 | 13 | 42 | 39 |
Pulp response | ||||
Yes | 6 | 26 | / | |
No | 2 | 9 | / | |
Unknown | 14 | 61 | / | |
AP | ||||
PAI = 1 | 15 | 65 | 36 | 33 |
PAI = 2 | 0 | 0 | 24 | 22 |
PAI ≥ 3 | 7 | 30 | 48 | 44 |
Root filling lenght | ||||
Adequate | / | / | 17 | 16 |
Short | / | / | 89 | 82 |
Long | / | / | 1 | 1 |
Mixed | / | / | 1 | 1 |
Root filling density | ||||
Adequate | / | / | 7 | 6 |
Voids | / | / | 101 | 94 |
Perforation | ||||
Absent | / | / | 106 | 98 |
Present | / | / | 2 | 2 |
Broken instrument | ||||
Absent | / | / | 93 | 86 |
Present | / | / | 15 | 14 |
Time since initial treatment | ||||
≤1 year | / | / | 21 | 19 |
>1 year | / | / | 35 | 32 |
unknown | / | / | 52 | 48 |
|
||||
Treatment sessions | ||||
Single-visit | 19 | 83 | 61 | 56 |
Multi-visit | 4 | 17 | 47 | 44 |
Root filling lenght | ||||
Adequate | 22 | 96 | 90 | 83 |
Short | 1 | 4 | 14 | 13 |
Long | 0 | 0 | 3 | 3 |
Mixed | 0 | 0 | 1 | 1 |
Root filling density | ||||
Adequate | 23 | 100 | 104 | 96 |
Voids | 0 | 4 | 4 | |
Complications | ||||
Perforation | 1 | 4 | 5 | 5 |
Broken instrument | 1 | 4 | 5 | 5 |
|
||||
Signs and symptoms | ||||
Absent | 22 | 96 | 98 | 91 |
Present | 1 | 4 | 10 | 9 |
AP | ||||
PAI = 1 | 23 | 100 | 89 | 82 |
PAI = 2 | 0 | 0 | 12 | 11 |
PAI ≥ 3 | 0 | 0 | 7 | 6 |
Coronal restoration | ||||
Temporary filling | 3 | 13 | 7 | 6 |
Permanent filling | 20 | 87 | 101 | 94 |
Post | ||||
Absent | 16 | 70 | 36 | 33 |
Present | 7 | 30 | 72 | 67 |
Restoration leakage | ||||
Good | 16 | 70 | 105 | 97 |
Poor | 7 | 30 | 3 | 3 |
Fracture | ||||
No | 23 | 100 | 106 | 98 |
Yes | 0 | 0 | 2 | 2 |
Included patients were seen during a consultation in the Dental Clinic of Bretonneau Hospital by the examiner. During this consultation, clinical and radiographic examinations were performed.
The extracted teeth were excluded from the analysis of results, because reason for extraction could not be determined. We included in the study the remaining included teeth at the consultation day (whether received initial treatment or retreatment).
The clinical and radiographic criteria were used for the results.
Endodontic success gathered “healed” teeth (absence of periapical radiolucency and no signs or symptoms) (Figure
Case classified as healed. (a) Preoperative view of the 37, painful, with failure of an initial endodontic treatment performed more than one year ago. Note the apical radiolucency (
Case classified as healing. (a) Preoperative X-ray of the 36. Patient referred for orthograde retreatment of 36, asymptomatic, with an apical chronic periodontitis on the mesial root. Note the apical radiolucency (
Case classified as failure. (a) Preoperative X-ray of the 46, painful, with an acute apical abcess. The previous endodontic treatment was performed 24 months ago. Note the apical radiolucency (
Healing was considered “uncertain” if the patient described signs or symptoms as the tooth looked healed on the periapical X-ray. In these situations, a CBCT was prescribed.Then, after CBCT examination, the tooth was classified as follows: if there was no lesion on the treated tooth but another affected tooth could explain the symptoms, the treated tooth was considered as “cured”; if the CBCT revealed a lesion on the treated tooth not visible on intraoral radiograph, it was recorded as “not cured”.
All teeth evaluated were selected for statistical analysis. The sample size is the number of teeth and not the number of patients.
We described the study population and we studied the bivariate associations between success and preoperative, intraoperative, and postoperative factors to identify potential outcome predictors. Fisher exact tests were conducted. All tests were interpreted at the 5% significance level. The number of failures was too small to have sufficient power to carry out multivariable analysis.
Of the 296 patients (363 teeth treated), 148 patients were categorized as lost: one died; 25 changed phone number; 104 did not call back after the message left on their answering machine; 11 did not answer; and 7 patients’ telephone numbers could not be retrieved from the database of the hospital.
148 patients (50% of the treated patients/of the eligible sample) were contacted: 7 patients had moved and could not come and 28 refused the follow-up visit (lack of availability, distance, and/or monitored regularly by a dentist). Ultimately, 113 patients were seen for visit. The recall rate is 38% (113 patients evaluated in 296 patients treated).
Sample of 113 patients and 139 teeth had been evaluated; 8 teeth extracted were excluded. Finally, 131 teeth were included in the study: 23 teeth had received an initial endodontic treatment and 108 teeth an orthograde retreatment.
The mean follow-up period was 35 months (
The average number of treatments per patient was 1.88.
13 women (68%) and 6 men (32%), 47% of patients, were less than 45 years and 53% of patients were over 45 years.
Among the initial endodontic treatment, we treated mostly posterior teeth (83%), multirooted teeth (70%), and maxillary teeth (61%).
13% of patients described preoperative clinical signs or symptoms and preoperative apical periodontitis that was observed in 30% of teeth.
There were 51 women (59%) and 35 men (41%), 47% of patients, less than 45 years old and 53% of patients were over 45 years. The population who received retreatment was uniform regardless of age or gender. Among the 108 endodontic retreatments, we treated mostly posterior teeth (96%), teeth multirooted (72%), and maxillary teeth (61%).
39% of patients had preoperative clinical signs or symptoms. Preoperative apical periodontitis was observed on 66% of teeth.
Silver cones were observed for three teeth and gutta-percha for all other teeth. The majority of endodontic treatment was inadequate: 84% of short length and/or overfilling, 94% with voids.
The global (initial and retreatment) success rate was 92% (121 teeth). Initial endodontic treatments resulted in 100% success (23 teeth): for this reason, the statistical treatment of this study was conducted only on retreatments. Among the 108 retreated teeth, 80% (86 teeth) was healed and 11% (12 teeth) was healing. 9% (10 teeth) was diseased. Among the diseased teeth, two were fractured (Table
Results.
Initial treatment | Retreatment | |||
---|---|---|---|---|
|
% |
|
% | |
Results* | ||||
Healed | 23 | 100 | 86 | 80 |
Healing | 0 | 0 | 12 | 11 |
Diseased | 0 | 0 | 10 | 9 |
Among the 42 teeth treated with preoperative signs and symptoms, the success rate was 83% (35 teeth), while for the 66 teeth without preoperative signs, there was 95% success rate. The success rate was statistically higher without preoperative signs and symptoms (
Of the 36 treated teeth without periapical radiolucency, 92% (33 teeth) was healed. Of the 72 teeth treated with preoperative AP, 90% (65 teeth) was healed or healing. But there was no statistically significant difference (Table
Analysis of preoperative factors.
Success | |||||||
---|---|---|---|---|---|---|---|
|
|
% |
|
OR | CI95% | ||
Signs and symptoms | |||||||
Absent | 66 | 63 | 95 | 1 | |||
Present | 42 | 35 | 83 | 0.04 | 0.24 | 0.06 | 0.98 |
AP | |||||||
PAI = 1 | 36 | 33 | 92 | 1 | |||
PAI = 2 | 24 | 21 | 87 | 0.64 | 0.12 | 3.45 | |
PAI > 3 | 48 | 44 | 92 | 0.83 | 1.00 | 0.21 | 4.78 |
Root filling length | |||||||
Adequate | 17 | 14 | 82 | 1 | |||
Short | 91 | 84 | 92 | 0.19 | 2.57 | 0.59 | 11.14 |
Root filling density | |||||||
Adequate | 7 | 4 | 57 | 1 | |||
Voids | 101 | 94 | 93 | 0.02 | 10.07 | 1.87 | 54.17 |
Complication* | |||||||
No | 92 | 84 | 91 | 1 | |||
Yes | 16 | 14 | 87 | 0.64 | 0.67 | 0.13 | 3.14 |
33 of the 36 teeth remained free of preoperative periapical radiolucency (94%) and 3 teeth developed apical periodontitis (6%).
Of the 24 teeth with preoperative PAI index = 2, 88% of periapical radiolucency disappeared (21 teeth), 8% of the lesions had the same PAI (2 teeth), and 4% of lesions increased in size (one tooth).
Of the 48 teeth with PAI index ≥3, the lesion size decreased for 92% (44 teeth) and remained stable for 8% (4 teeth).
Of the 72 teeth with AP, a higher success rate was obtained when the preoperative root filling was short (92%) versus adequate (83%). It was not significantly different (
Among 101 teeth with root filling voids, success rate was 94% and 57% if no void was recorded (7 teeth). The difference was statistically significant (
(i) Success rate was not different according to the preoperative presence of a broken instrument. Of 108 retreated teeth, 15 teeth had a preoperative broken instrument with a success rate of 87%. The 93 remaining teeth without broken instrument had a success rate of 91%.
(ii) Of the 108 retreated teeth, the two teeth with preoperative perforation healed.
(iii) In short, preoperative outcome predictors were signs and symptoms (absent, 95%: present 83%;
The other preoperative factors did not seem to be outcome predictors in this study.
Of the 61 teeth treated in one session, the success rate was 89% while for the 47 teeth treated in two sessions, a success rate of 94% was obtained. There was no significant difference. However, the analysis of 72 teeth with preoperative periapical lesion gives a better result on the treated teeth in two sessions (97%) compared to those treated in one (86%).
Among the 89 teeth with preoperative short root canal filling, adequate length was found for 72 teeth (81%).
Of the 10 teeth treated with intraoperative complications (perforation or fractured instrument), there was no failure.
No statistical difference was demonstrated relative to the presence of a post or to the quality of the coronary restoration.
91% of teeth (98 teeth) were asymptomatic the day of the visit; 9% of teeth (10 teeth) had clinical signs or symptoms. For the 10 symptomatic teeth, 4 teeth were considered not healed and 6 teeth as uncertain. A CBCT was prescribed to patients whose conclusion was uncertain. After viewing the 3D examination, two of them have been added to the successes and 4 to failures.
In this retrospective study, the outcome of the endodontic treatment was assessed on both clinical and radiographic criteria, as recommended by the European Society of Endodontology [
A new classification based on effectiveness/ineffectiveness was recently proposed [
In this study, effectiveness of initial treatment raised to 100% and to 91% for retreatment.
For initial treatment, success rate was 100% both for vital teeth (pulpitis) or nonvital teeth (necrosis and apical periodontitis). Previous study reported that outcome of endodontic treatment is different between vital and nonvital cases. Unfortunately this study cannot answer the question, due to a too small sample size that has statistical significance (only 23 teeth).
Healed teeth (80%) (see Figure
In this study, the success rate decreases by 5% over 4 years. The success rate is 92% 1 year after completion of endodontic treatment, 90% after 2 years, 91% after 3 years, and 87% after 4 years (Figure
Diagram of success rate of the follow-up per year.
Radiographic examinations allow the evaluation of the main success criterion and the presence or absence of a radiolucent image. 2D imaging underestimates the presence and the volume of AP and this is why a PAI score = 1 was chosen as an AP free indicator [
Periapical lesions were scored with the PAI index [
Contribution of the CBCT in the decision making process. (a) Periapical preoperative X-ray. Patient referred for the endodontic retreatment of 25. Note the periapical radiolucency on 25. (b) Control 12 months after the retreatment of 25. There are symptoms and persistent radiolucency on 25. The tooth was classified as “failure”. (c) CBCT coronal and sagittal slides of 25: a periapical surgery is indicated, (d) X-ray control 12 months after apical surgery and apical retrofilling with Biodentine (Septodont) which has the same radio-opacity as the root dentin.
The presence or absence of an AP significantly affects the rate of success of endodontic treatment [
The length of preoperative root canal treatment was more than 2 mm from the apex on 82% of the retreated teeth (89 teeth), which is described as insufficient. A retrospective study in a German dental school has confirmed the negative impact of insufficient length on the success of endodontic treatment [
When the preoperative length was correct, the rate of success of our treatments was lower: 83%. It is probably the sign of a virulent infection less sensitive to retreatment procedures [
For the 2 teeth with preoperative periapical overfilling of gutta-percha we obtained healing after retreatment. During the retreatment overfilling of gutta-percha that occurred in three cases, two teeth healed and one was healing. Among the teeth classified as failure, three are with extrusion of sealer. Among the teeth classified as healing four are with this kind of extrusion. The majority of teeth with sealer extrusion (21/28) are classified as healed. It’s accepted that it is better to avoid overfilling but a controlled overfilling (“puff”) is not a predictor of failure.
Perforations were long considered as major complications reducing the prognosis for survival of the tooth. Treated teeth from the Toronto Study have a better healing rate if the tooth is not perforated (86% versus 36%) [
Successful endodontic retreatment of 36 symptomatic, with root perforation. (a) Preoperative X-ray: note the lateral periodontitis regarding the mesial perforation of the mesial root. (b) Peroperative X-ray: the perforation was filled with MTA and the root canal by vertical condensation of warm gutta-percha. (c) Follow-up for 1 year: tooth classified as “healed” (no symptoms, no radiolucency).
The impact of a fractured instrument in a root canal must be assessed accurately. According to a recent systematic review [
Successful endodontic retreatment of 27 symptomatic, with broken instrument. (a) Preoperative X-ray of a patient sent for orthograde retreatment of 27, painful, with broken instrument (Lentulo) in the mesio-buccal root. The initial treatment was recent. (b) Peroperative X-ray: the broken instrument has been removed and apical patency was found before the cleaning, shaping, and filling of the canal. (c) Recall X-Ray with follow-up for 4 years. Note that lateral radiolucency had disappeared. The tooth is classified as “healed”: absence of symptoms and periodontal healing.
Fractured file does significantly reduce healing of the lesion in the presence of AP. Therefore, although it is recommended that file removal should be attempted when possible, this does not appear to be evidence-based in the absence of apical disease [
For the “2-session” group, with a Calcium Hydroxide intersession dressing, a high rate of success was obtained (94%), while it was lower for treatment in one session (89%). In the presence of an AP, the probability of healing is here better for the multisession versus single session treatment (97% versus 86%) but the
The quality of the coronary filling is an important factor in the prognosis of the treated teeth [
The presence of a post into the root canal is a frequently discussed factor. Our study demonstrates a rate of success of 89% for teeth with a post and 94% for those restored without post. The choice of a post should be considered a last-ditch attempt, and bacterial contamination should be avoided.
None of the referred patients presented any contraindication relative to endodontics. One patient had a specific disease (Hodgkin lymphoma). The tooth of this patient was extracted due to a root fracture.
Despite the limited sample size and the questionable recall rate, the results of this study confirm partially data from previous studies. It confirms the importance of identified predictors, including the initial symptoms and quality of initial treatment, as significant factors in the prognosis of the treatment. In the sample of patients involved, the presence of periapical radiolucency, the number of sessions, and the quality of the coronary restoration were not identified as statistically significant predictors.
The study investigates other factors such as the intraoperative incidents (broken instrument, perforations). However, for these factors due to the limited number of cases, the relative importance of each is not statistically significant. A larger sample size is needed to assess all outcome predictors of endodontics treatment more precisely.
Within the limits of this retrospective study, this work highlights the reliability of the initial endodontic treatment and the strong potential of endodontic retreatment when performed by trained and competent practitioners. The effectiveness of initial treatment is maximal and remains very high for retreatment.
The authors declare that there is no conflict of interests regarding the publication of this paper.