Clinical and radiological analysis of the causes for endodontic treatment failure

Summary Introduction Development of inflammatory lesions or their persistence after primary treatment is considered endodontic failure. The reason for failure can be complex anatomy of the canal system and numerous iatrogenic factors. The objective of this study was to analyze, clinically and radiographically, the causes of primary endodontic treatment failure and assess possibilities for retreatment of teeth with failed endodontic treatment. Method The study included 79 teeth (36 multirooted and 43 singlerooted tooth) indicated for repeated endodontic treatment. Based on the radiographic assessment of the status of periapical structures, teeth were divided into two groups. The first group included teeth without periapical lesions, i.e. the healthy periodontal tissues (PAI score of 1 and 2) in which retreatment was required for prosthodontic reason due to the poor quality of obturation (28 teeth), and the second group included teeth with visible signs of periapical tissue damage (PAI scores 3, 4 and 5) (51 teeth). In both groups, quality of obturation, coronal sealing and the presence or absence of clinical symptoms was analyzed. Results The most common radiographic finding of definitive obturation was short filling (65.8% of cases); “forgotten” canals (25.3%); non-homogeneous obturation with correct length (5.1%) and fractured instrument (3.8%). There was significant difference between healthy periodontal ligament and adequate restoration (P < 0.001). In 95% of patients with symptoms, changes in the periapical tissue were observed. Also, there was significant difference in the presence of symptoms after primary treatments, between the teeth with healthy apical periodontal tissue and teeth with periapical lesions (P = 0.019). Conclusion The outcome of the root canal treatment is significantly affected by the quality (density) of obturation and the presence and quality of coronal restoration. In patients with symptoms there were changes in the periapical tissue.


INTRODUCTION
Healing of periapical lesions presented as reduction or disappearance of existing periapical radiolucency is expected after adequately conducted endodontic treatment. However, there are situations when bone repair is absent and there is no reduction in periapical radiolucency, often qualified as endodontic failure [1]. The success rate of endodontic treatment ranges from 53% -97% [2][3][4][5]. Higher percentage indicates that almost every endodontic treatment is successful, while lower limit interval suggests that every other is unsuccessful.
Despite the undoubted frequency in everyday clinical practice, there is still no exact definition of endodontic treatment failure. Many clinicians came to the consent that lack of pain and other clinical symptoms or maintained function of endodontically treated teeth are important parameters of successful endodontic treatment [6]. The failure of endodontic treatment includes radiographic appearance of inflammatory lesions in the periapical tissue, which had not existed before or persistence of or enlargement of the radiolucency after undertaken primary treatment [6,7].
Sometimes failure can occur even if endodontic treatment was properly managed and all procedures are fully respected. The reason for this is the complex anatomy of the canal system and numerous ramifications and anastomoses between the main and accessory canals that cannot be adequately treated or obturated using contemporary instruments, materials and techniques. Noninstrumented region of endodontic space can contain bacteria and necrotic tissue, even when not visible on the X-ray [12,24].
There are factors outside the root canal, within the inflamed periapical tissue that could adversely affect post-operative healing of periapical lesions. Persistence of asymptomatic periapical radiolucency after thorough endodontic therapy can be caused by extra-radicular in-fection, true cysts, foreign bodies, the presence of cholesterol crystals or scar healing of the tissue [25].
The aim of this study was to analyze, clinically and radiographically, the causes of primary endodontic treatment failure and assess possibilities for retreatment of teeth with failed endodontic treatment.

METHOD
The study was conducted at the Clinic of Restorative Odontology and Endodontics, Faculty of Dental Medicine, University of Belgrade, Serbia. All participants signed consent to voluntary participation in the study after introducing with objectives and expected outcomes of the research.
67 patients of both genders, aged 24-79 years, and 79 teeth (36 multirooted and 43 singlerooted tooth) indicated for repeated endodontic treatment were included in the study. One operator carried out clinical trial, while two researchers interpreted the assessment of radiographic outcome.
All 79 teeth had inadequate radiographic obturation and that was the key criterion in the assessment of the failure of endodontic treatment. 52 teeth had short filling while 4 teeth had non-homogeneous filling. "Forgotten" canals were found in 20 teeth while 3 teeth had separated instruments (Table 1). In 36 teeth restoration was adequate or had valid prosthetic restorations, 5 teeth were without fillings a longer period of time, and 43 teeth had inadequate restoration.
Periapical status of each tooth prior to the re-treatment was assessed radiographically using PAI (periapical index) system as follows [26]: 1-PAI normal periapical structures 2-PAI small changes in bone structures that is not pathognomonic for apical periodontitis 3-PAI changes in bone structure with decalcification, characteristic for apical periodontitis 4-PAI periodontitis with clearly defined zone of radiolucency 5-PAI advanced periodontitis with signs of exacerbation and expansion of bone.
PAI score was determined for each tooth individually. X rays were analyzed on the light box using magnifying lens. Multiroooted teeth were evaluated according to the maximum damage of periodontal structures in any of the roots. Based on the state of periapical structures, teeth  1  14  10  4  12  2  2  14  10  4  9  5  3  29  16  3  1  9  9  20  4  15  11  1  2  1  3  12  5  7  5  2  3  4  ∑  79  52  4  3  20 36 43 were divided into two groups The first group included teeth without periapical changes (PAI score 1 and 2) where retreatment was necessary for prosthetic reasons and poor quality of definitive obturation (28 teeth). The second group included teeth with visible signs of periapical tissue damage (PAI scores 3, 4 and 5) and included 51 teeth. The second parameter in the analysis was the existence of clinical symptoms after the initial treatment. The first group included cases without clinical symptoms (diagnosed as incidental findings), and the second group included teeth with present clinical symptoms: pain, swelling, sensitivity to percussion, present sinus tract and others ( Table 2). Figures 1-4 and Tables 3-6.

Results are presented in
Periodontal tissues were found healthy (PAI 1 and 2) in 35.4% of cases, while some changes in periapical tissue (PAI 3, 4 and 5) were recorded in 64.6% of cases. Short obturation was recorded in 65.8% of cases, while "forgotten" canals that were detected in 25.3% of teeth. Non-homogenous filling with correct length was noted in 5.1% of teeth and fractured instrument was found in the root canal of 3.8% of analyzed teeth ( Table 3).
The quality of the coronal seal was inadequate in 54.5% of cases, while in 45.5% of cases coronal restorations had satisfactory quality ( Table 4). Most of teeth with healthy    (Table 5).
χ2 test showed high association between healthy periodontal tissue and adequate restoration on one side and micro leakage of inadequate restoration and periapical changes (P <0.001) ( Table 5) In regards to the symptoms, 24% of patients had persistence of the symptoms after the initial endodontic therapy, while the remaining 76% of cases were asymptomatic. Out of all asymptomatic patients 95% had some changes in the periapical tissue. Fisher's exact test showed association between the presence of symptoms after initial treatments and teeth with periapical lesions (P = 0.019) ( Table 6).

DISCUSSION
This clinical study was conducted with the aim to analyze the outcomes and causes of the failure of primary endodontic treatments and to facilitate planning for clinicians how to perform retreatment. All cases were selected from everyday clinical practice at the Clinic of Restorative Odontology and Endodontics, School of Dental Medicine, University of Belgrade. A single therapist carried out retreatments.
Consent to participate in the study was signed by 37 patients of both genders, aged 24 to 79 years. Based on the clinical and radiographic examination they required repeated endodontic therapy. The study did not include patients with general diseases and those taking antibiotics in the last 3 months, and teeth with extensive decay destruction or poor periodontal status.
Primary endodontic treatment was done in 2 cases within a year, in 17 cases (34.7%) 1-5 years before the diagnosis of failure of previous treatment, while in 30 teeth (61%) primary endodontic treatment was done more than 5 years ago. Endodontically treated teeth have long survival rate in general. In fact, over 60% of root canal treated teeth are functional for more than 5 years. Salehrabi and   Rotstein (2004) carried out an extensive epidemiological study in the United States on 1,462,936 teeth. After 8 years, 97% of teeth were still present in the oral cavity [4]. On the other hand, other epidemiological studies [27,28,29] in the recent years indicated that over 30% of endodonticcally treated teeth were diagnosed with chronic periapical lesions (apical periodontitis) or according to Friedman-"post-treatment endodontic disease" [1]. From a total of 79 teeth, 35% of the teeth had no visible signs of periapical bone destruction, while in 65% of teeth the presence of chronic periapical lesions was registered. PCR analysis of the samples taken from the root canals of these teeth after removing old canal filling identified microorganisms in all teeth that had changes in the periapical tissues. The most frequently identified microorganism, E. faecalis, was detected in 94% of the root canals with chronic periapical lesions.
Only 24% of patients had clinical symptoms such as pain, swelling, the existence of a sinus tract, or sensitivity to percussion and pain on biting. Most prevalent symptoms were sensitivity to percussion (59%) as a sign of chronic inflammation of the periapical tissue, and pain (47% of cases) as a sign of acute exacerbation of chronic periapical lesion.
All 79 tooth treated in our study had inadequate obturation. Most frequently registered was short filling in 65% of cases, "forgotten canals" (25%), clinically non-homogeneous filling (5.1%) and the presence of fractured instruments in 3.8% of canals. Whether inadequately obturated root canal is going to cause the failure of endodontic treatment depends primarily on the presence of bacteria in the root canal. If the vital pulp was treated and coronal restoration properly sealed, impermeable to bacteria, changes in the periradicular tissue will most likely not occur. However, if the canal was infected, an empty space in the apical part of the root will probably cause persistent intraradicular infection or maintain periradicular inflammation after completion of endodontic therapy [8][9][10][11][12][13][14][15].
In addition to the length of the apical canal filling, density ie. hermetic canal filling is an important factor for successful endodontic treatment. Unfortunately, micro computed tomographic studies have shown that even the most modern materials and techniques of instrumentation and obturation are not able to obturate root canal non-porously or to be impermeable to bacteria. . After scanning and measurement of the volume of voids and unfilled space in obturated root canals, with respect to their total volume, they came to the con-clusion that no technique of obturation provides absolute hermetic sealing of the endodontic space. Furthermore, the difference in the percentage of empty space between the novel adhesive endodontic materials and gutta-percha as gold standard was not statistically significant [30,31].
It is obvious that the quality of obturation affects the outcome of endodontic treatment but this is not the only prerequisite for success. In our study all patients with inadequate obturation were referred to endodontic retreatment, even though 17.7% of teeth did not have any changes in the periapical tissues (PAI 1), while at 17.7% of teeth there was slightly enlarged periodontal membrane (which is not pathognomonic finding of apical periodontitits) (PAI 2). In these cases, the retreatment was needed due to prosthetic reconstruction (intraradicular post). Such teeth were adequately restored in 75% of cases (21 out of 28 teeth) that provided good coronal seal, reduced microleakage and prevented (re) infection of the root canal and periapical tissues. Only one tooth with healthy periapical tissue did not have an adequate restoration and showed signs of acute infection (not visible changes on the x ray). Ray and Trope demonstrated that defective coronal restoration and adequate obturation have higher percentage of failures compared to the teeth with appropriate crown restoration and inadequate obturation [23]. Only 9% of teeth with adequate coronal filling and root canal obturation showed failure, in contrast to the teeth where obturation and coronal restoration were defective where the ratio was 82%. Gillen et al. (2011) conducted a systematic review of available literature about the effect of coronal restoration and root canal obturation and concluded that success of endodontic treatment would be higher if both, endodontic treatment and coronal restoration were done properly [21].
Out of all teeth with changes in the periapical tissue, 84% had inadequate restoration, and 50% of them had symptoms indicating the importance of good marginal seal. Similar findings were reported by Liang et al. (2011)  who used periapical radiography and CBCT to analyze factors required for successful endodontic treatment. Two years after pulpectomy, periapical radiography showed the presence of periapical lesions in 12.6%, while CBCT detected two times more chronic periapical changes (25.9%). Interestingly, 80% of root canal fillings radiographically determined as "short", on CBCT were up to the apical terminus. They also concluded that density and apical extension of the root canal obturation significantly affected the outcome of endodontic treatment as judged by periapical radiographs. By analyzing data obtained by CBCT, for the success of the root canal treatment critical factors were density of the root canal filling and the quality of the coronal restoration [22]. Teeth with healthy periodontal tissue were sent to endodontic retreatment due to radiographically inadequate obturation (part of the preparation for prosthodontic rehabilitation) or were detected as incidental finding. Dilemma of whether teeth with inadequate obturation, healthy periapical tissue and without clinical signs and symptoms, should be retreated, is always current among endodontists. If tooth has already adequate restoration, it can be monitored by regular check-ups. However, if the tooth is planned to support fixed prosthodontic construction, with intraradicalar post, retreatment is strongly recommended [32]. Success of repeated endodontic treatment relies on: proper diagnosis of the endodontic failure (periapical radiography, CBCT), adequate desopturation and repeated cleaning and shaping of the canal (the expertise of the therapist, instruments and materials), highquality three-dimensional hermetic obturation of the root canal system, and timely and adhesive restoration placement after retreatment is finished.

CONCLUSION
The outcome of the root canal treatment is significantly affected by the quality (density) of root canal obturation and the presence and good quality of the coronal restoration. Most teeth with healthy periodontal tissue at the time of diagnosis of the failure of primary endodontic treatment were adequately restored. However, most teeth with changes in the periapical tissue had inadequate coronal restoration. Teeth in patients who had symptoms mostly had visible changes in the periapical tissue.