Abstract

Background: The success rate of contemporary apical surgery is high after the introduction of magnification, ultrasonics, and bioceramic materials. Persistent apical periodontitis following surgical treatment is usually related to the failure in eliminating or at least sealing residual bacteria occurring in the root canal system.

Case Description: This article reports on a clinical case of postsurgical apical periodontitis associated with a mandibular molar exhibiting failing root-end filling with a bioceramic (calcium silicate-based) material. The tooth had been subjected to retreatment and later to apical surgery because of persistent apical periodontitis. After a period in that the lesions had apparently healed based on radiographs, the appearance of a sinus tract and the observation of recurrence of the apical periodontitis lesions on both roots as shown by CBCT led to indication for a new surgery. Surgery revealed a fracture in the mesial root. The tooth was extracted and the distal root with the associated apical periodontitis lesion was processed for histobacteriological analysis of the possible etiology of postsurgical disease. A residual bacterial infection, favoured by substandard technical surgical issues, was the most likely cause of the periapical inflammation. It was also apparent that root end filling was not placed along the long axis of canal to depth of at least 3mm, and the bioceramic material used as root-end filling allowed bacteria to leak across and grow on it forming biofilm structures.

Conclusion: The observation of bacteria leaking through the bioceramic material used as root-end filling questions the reported antimicrobial properties of bioceramic materials.

Introduction

The success rate of apical surgery is relatively high, especially after the introduction of magnification, ultrasonics for root-end resection and preparation, and bioceramic (also referred to as hydraulic, calcium silicate-based, or MTA-based) root-end filling materials (Rubinstein & Kim 2002, Setzer et al. 2010, Tsesis et al. 2013, von Arx et al. 2019). Persistent apical periodontitis following surgical treatment is usually related to the failure in eliminating or at least sealing residual bacteria occurring in the root canal system (Hsu & Kim 1997, Ricucci & Siqueira 2008). An appropriate sealing by the root-end filling material can preclude intracanal bacteria to have access to nutrients from the periradicular tissues in the form of tissue fluids and inflammatory exudate (Antunes et al. 2016). Furthermore, if the material has some antibacterial properties, it can help control the residual infection (Chong et al. 1994).

For residual bacteria to cause postsurgical apical periodontitis, they have to establish a frank contact with the periradicular tissues, which may occur when they are located in apical ramifications, isthmuses, and infected dentinal tubules exposed after root-end resection and/or made patent after cementum resorption; or because of a poor sealing provided by the root-end filling material, or a root crack/fracture that communicates the periradicular tissues with the root canal or the gingival sulcus (Friedman 2008, Ricucci & Siqueira 2008, Siqueira & Rôças 2022). A persistent intraradicular infection can be regarded as the main cause of postsurgical apical periodontitis.

Bioceramic cements have been widely used as root-end filling material with a highly reported success rates (Kohli et al. 2018, Torabinejad et al. 2018). These materials have been extensively investigated and many formulations are currently available (Prati & Gandolfi 2015, Primus et al. 2019). Bioceramic materials are hydraulic and hygroscopic and exhibit important properties, including sealing ability, biocompatibility, antimicrobial activity, and bioactivity (Camilleri et al. 2022).

This article reports on a clinical case of postsurgical apical periodontitis associated with a tooth exhibiting a failing root-end filling with a bioceramic material. A persistent bacterial infection somewhat favored by technical issues was the most likely cause of the postsurgical disease.

Case report

A 61-year-old African American male was referred to postgraduate endodontic clinic to evaluate his mandibular right first molar because of discomfort to chewing. Medical history revealed that the patient had hypertension, poorly controlled diabetes, and end-stage renal disease (ASA III). The tooth had been root canal treated and restored by post/core and crown approximately one and a half years before. The patient had just started orthodontic treatment. Extra-oral and intra-oral examinations revealed healthy soft tissue. The tooth was slightly tender to vertical percussion. Probing depths were not exceeding 3 mm.

Periapical radiographs showed that short root canal fillings and periapical radiolucencies were present at the mesial and distal roots of tooth #30 (Fig. 1 A). The diagnosis of failing root canal treatment was made and the patient was given all treatment options, including endodontic retreatment, apical microsurgery and extraction. He chose to go through non-surgical retreatment and an informed consent was obtained.

All retreatment procedures were performed under an operating microscope and rubber dam isolation. A through crown accessing was made, and the screw post was removed with ultrasonic tips from the distal canal. Root canal filling materials were removed with chloroform and H-files. The mesio-lingual canal had no patency. The root canals were instrumented with EndoSequence rotary files (Brasseler, Savannah, GA). Mesial canals were prepared up to size #35 and the distal canal up to size #50. Irrigation was conducted with 4% NaOCl during preparation, followed by 17% EDTA and 2% chlorhexidine after preparation was complete. Calcium hydroxide mixed with sterile saline to a creamy consistency was placed as an interappointment intracanal medicament.

The patient returned two weeks later, asymptomatic, and the canals were filled with a sealer-based filling technique using BC Sealer (EndoSequence BC Sealer, Brasseler) and gutta-percha cones (EndoSequence BC points, Brasseler).

A flowable composite was placed as an orifice barrier (PermaFlo, Ultradent, South Jordan, UT). The access cavity was closed with glass ionomer (Fuji IX, GC America, Chicago, IL). A final radiograph was taken after retreatment (Fig. 1 B).

Six months later, the patient returned for follow-up examination. The tooth had been asymptomatic. Core build-up and a temporary crown were placed. A periapical radiograph showed that the mesial and distal radiolucencies had slightly decreased in size (Fig. 1 C). At the 12-month follow-up visit, the tooth was asymptomatic. A temporary crown was still present because the patient decided to postpone the permanent restoration. A periapical radiograph revealed that the radiolucencies had remained the same size (Fig. 1 D). A CBCT scan was taken at this follow-up. This confirmed the size of the apical periodontitis lesions (Fig. 1 E), and revealed the presence of a missed buccal canal in the distal root (Figs. 1 F-G). This canal seemed to originate from the treated canal, approximately at the transition from the coronal to middle third (Fig. 1 G). The case was regarded as a failure, and the patient was once again presented with the options of retreatment, apical microsurgery, or extraction. During discussion, the patient was informed about the poor prognosis of an orthograde retreatment. The possibility to negotiate the complex root canal anatomy in the distal root and to bypass the blockage in the mesiolingual canal was realistically considered negligible. The patient opted for apical microsurgery and gave informed consent. He was warned about the necessity of keeping blood pressure and glucose under control before surgical intervention.

Surgery was performed six months later for patient convenience. After mouth rinsing with 0.12% chlorhexidine gluconate for 60 seconds, local anesthesia was given using 1.7 mL 2% lidocaine with 1:100,000 epinephrine for inferior alveolar nerve block and 1.7 mL 2% lidocaine with 1:50,000 epinephrine for buccal infiltrations. After flap elevation, osteotomy and root resection were made. Methylene blue was applied on the root, but no fracture line was noted. Three millimeters of each root was resected and stained again. A middle mesial and a distal buccal canal were noted; 3 mm root-end preparations were filled with BC putty (Brasseler) (Figs. 1 H-K). The flap was sutured without placing any grafting material. A postoperative radiograph was taken (Fig. 2 A). Tylenol and Peridex were prescribed for pain and inflammation control at the end of the appointment.

Seven days after surgery, the patient returned for suture removal. He reported no discomfort from the surgical area. A follow-up radiograph taken after 8 months showed complete bone healing (Fig. 2 B). In the meantime, the tooth had been restored with a gold crown (Fig. 2 B).

Forty-six months after surgery, the patient was referred back to the clinic because a buccal sinus tract was observed. Periodontal probing gave depths of 3 mm circumferentially, except in the mesiobuccal area, where the probing depth exceeded 10 mm. A periapical radiograph revealed periapical radiolucencies on both roots, and an osteolytic lesion interradicularly (Fig. 2 C). A gutta-percha point inserted into the sinus tract pointed to the mesial root (Fig. 2 D). CBCT exam was indicated and sagittal scans revealed the presence of periapical lesions on both roots (Fig. 2 E). Axial scans revealed a V-shaped bone dehiscence buccally along the whole mesial root, suggestive of vertical root fracture (Fig. 2 F). Coronal scan of the distal root confirmed the apical radiolucency, but showed evidence of healing of the buccal cortical bone (Fig. 2 G).

Resurgery was proposed. The patient agreed and understood the risks of this procedure and that the prognosis would be hopeless if fracture lines were noted on the root during surgery. Informed consent was obtained. After flap elevation, bone fenestration was noted at the mesial root, with a fracture line present (Fig. 2 H). Before tooth extraction, a bone window was made with piezotome in the cortical bone covering the distal root (Fig. 2 I) and the apical soft tissue exposed (Fig. 2 J). The apical third was resected (Fig. 2 J) and the soft tissue, with the consistency of granulation tissue, curetted (Fig. 2 K). The soft tissue lesion, the cortical bone fragment, and the resected apex of the distal root were immediately immersed in fixative for histological processing. The patient gave consent for analysis. The same post-op instruction was given to the patient.

Tissue processing

Immediately after surgery, the biopsy specimens were immersed in 10% neutral buffered formalin for 48 hours. After fixation, the soft tissue was dehydrated, cleared, and embedded in paraffin. Serial sections were cut (Fig. 3 A-B). The apical segment and the cortical bone were demineralized in a solution consisting of 22.5% (vol/vol) formic acid and 10% (wt/vol) sodium citrate for 4 weeks. The end-point was determined radiographically. At the end of the demineralization process, the specimens were washed in running water for 24 hours and dehydrated in ascending grades of ethanol. Clearing in xylene followed.

Photographs of the cleared distal apex were taken while the specimen was immersed in xylene (Figs. 3 C-D). The apex was oriented in the paraffin block, so that longitudinal sections could be cut on a mesio-distal plane. With the microtome set at 4–5 μm, meticulous serial sections were taken, starting from the buccal aspect, until the specimen was exhausted. Sections were stained with Hematoxylin & Eosin, and a modified Brown and Brenn technique for staining bacteria (Taylor 1966).

The stained sections were examined using a light microscope (Leica DMLB; Leica Microsystems, Wetzlar, Germany), and photomicrographs were taken (Leica DFC420; Leica Microsystems, Wetzlar, Germany).

Histopathologic and histobacteriologic observations

Sections of the periapical soft pathologic tissue revealed that the lesion was a non-epithelialized granuloma (Figs. 3 A-B). Areas with prevalence of collagen bundles and scattered inflammatory cells were seen alternated with areas exhibiting severe concentrations of acute (polymorphonuclear leukocytes) and chronic (lymphocytes, plasma cells, macrophages) inflammatory cells, with congested vessels (Fig. 3 B).

For the distal root apex specimen, serial sectioning commenced at the buccal side and proceeded in the lingual direction. The first sections of the series, cut on the buccal aspect and not encompassing the retrograde filling, showed that the dentin on the resected apical surface was covered with a layer of newly formed stratified cellular cementum (Fig. 3 E). This appeared detached from the underlying dentin in some areas, where small amounts of bioceramic material could be observed (Fig. 3 F).

This layer extended to the mesial and distal apical contour, so that the apical structure was wrapped by an uninterrupted sheet of eosinophilic calcified material. This newly formed external layer could be easily distinguished from the pre-existing cementum, which was less eosinophilic (Fig. 3 E).

Serial sections cut through the buccal root-end cavity revealed that the preparation was centered in the buccal canal (Fig. 4 A), but it soon deviated from the long axis of the canal (Figs. 4 B-D). The unprepared canal space, coronal to the retrofilling, was densely packed with necrotic debris and bacterial biofilms (Fig. 4 C).

The mass of bioceramic material filling the root-end cavity appeared to be heavily infected. Bacterial aggregations were observed, colonizing voids and fissures throughout the entire extent of the root-end filling (Figs. 4 A-E). The newly formed cementum layer could be observed on the resected dentin surface, but it was absent adjacent to the filling material (Fig. 4 A, C-D). Thick bacterial biofilms were present on the external surface of the cementum layer, and on the bioceramic material (extraradicular biofilms). Branching ramifying bacterial forms with varying amounts of extraradicular polymeric structure were predominant in these extraradicular biofilms (Fig. 4 E-G).

Sections cut approximately at the center of the distal root apex, through the isthmus, revealed that the root-end cavity was poorly extended in a coronal direction. The overlying isthmus was clogged with necrotic debris and biofilms (Fig. 5 A).

Histologic sections encompassing the distolingual canal showed that the retrofilling bioceramic material was in contact with the root canal filling material (Fig. 5 B). No bacteria could be observed in the previously treated root canal space (Fig. 5 C), but they were abundant in the bioceramic material, as well as extraradicularly (Fig. 5 D).

DISCUSSION

This article reports on a case of postsurgical apical periodontitis in a mandibular molar, which had to be extracted because of a vertical fracture in the mesial root observed at resurgery. Radiolucent lesions were present on both roots. Because the distal root had its own individual lesion and no sign of fracture, it was histopathologically and histobacteriologically evaluated for the possible cause of postsurgical disease.

The presence of a newly formed layer of cementum “wrapping” the resected apex in areas adjacent to the root-end cavity indicates that there was an initial healing response after the first surgery. This cementum, of the cellular type, was juxtaposed to the resected dentin surface and continuous with the layer formed over the preexisting cementum on the mesial and distal aspects. The presence of small amounts of bioceramic material interspersed between dentin and cementum seemed not to have interfered with cementum deposition. Actually, animal studies have demonstrated that the use of bioceramic materials for root-end fillings can permit or stimulate new hard tissue formation at the site of the resected root (Economides et al. 2003, Tawil et al. 2009, Walivaara et al. 2012). This property was confirmed in humans, where newly formed cellular cementum and uninflamed connective tissue were observed in contact with a calcium-silicate material used as a retrofilling material (Ricucci et al. 2020). Furthermore, the presence of an intact buccal cortical bone, repairing the osteotomy site, confirms the initial healing response. In fact, radiographs taken at 8-month follow-up suggested that healing had taken place on both roots. However, the lesions apparently recurred at a later follow-up, when the tooth also developed a vertical fracture in the mesial root with sinus tract formation. Both radiographs and CBCT revealed the presence of radiolucent lesions on both roots. Late failures have been reported to be more frequent after surgery than retreatment (Torabinejad et al. 2009). The most likely cause of the recurrent postsurgical lesion reported here was a persistent intraradicular bacterial infection.

A heavy bacterial infection, in many places organized as biofilm structures, was observed in the untreated canal as well as throughout the filling material and also reaching the extraradicular compartment. The possibility exists that leakage of bacteria through the bioceramic material used as root-end filling occurred over the years and was the cause of the postsurgical inflammatory disease. The fact that the root-end cavity was shallow in the area of the isthmus may be indicative of a substandard surgical treatment. It is commonly recommended that the root-end cavities be unified, to include the isthmus, with a depth of at least 3 mm to accommodate a minimum amount of filling material to provide a good seal (Kim & Kratchman 2006). However, when performing surgical preparation of an isthmus, there is a balance one must consider between fracturing/perforating a tooth with a full 3-mm ultrasonic preparation versus the risk of a compromised seal. It is also salient to note that long ultrasonic tips (up to 9 mm in length) can be used for preparation of missed canals or canals with very short orthograde fills, but in many cases it can be technically difficult to remain within the long axis of the canal.

It was apparent that the bioceramic root-end filling failed to prevent bacterial leakage, as biofilm structures were seen even along the material mass. This indicates that not only was the sealing inadequate, but the reported antimicrobial properties of bioceramic materials (Lovato & Sedgley 2011, Koutroulis et al. 2023) were not sufficient to prevent both the bacterial advance through fissures and voids in the material mass and biofilm formation on the bioceramic material itself. It is also possible that the residual bacterial biofilms in the untreated canal may have been displaced and mixed with the root-end filling material during its placement, but this also points to antimicrobial ineffectiveness of the material. Most of the antimicrobial effects attributed to bioceramic materials are probably related to calcium hydroxide formation during setting (Camilleri et al. 2005, Prati & Gandolfi 2015). The pH-dependent effects of calcium hydroxide have limitations, as they are rapidly inactivated when in direct contact with tissue fluids, which are rich in proteins and other buffering systems that may reduce or render calcium hydroxide innefective (Siqueira & Lopes 1999). Moreover, the antimicrobial effects of bioceramic materials have been shown to be significantly reduced after interaction with blood (Farrugia et al. 2017), which results in the formation of surface deposits that reduce the pH.

In a case recently reported, the presence of bacteria coronally to the root-end filling was not able to affect the periapical connective tissue, which was uninflamed (Ricucci et al. 2020). It is difficult to compare the opposite results between these cases, but the technical standards of the root-end fillings may have accounted for the differences observed. In addition, in the present case, the amount of residual intracanal infection was much larger than that seen in the previous case.

The presence of an isthmus is a risk factor for postsurgical disease (Hsu & Kim 1997, von Arx 2005). There are different types of isthmuses (Hsu & Kim 1997). Weller et al. (Weller et al. 1995) distinguished the complete from the incomplete isthmus. The management of a complete isthmus is relatively easy with the use of appropriate ultrasonic instruments; however, the incomplete isthmus requires a careful approach with thin ultrasonic tips troughing along it (Carr 1997). The aim of the root-end preparation is to remove the intracanal filling material and the necrotic/infected debris and create a cavity that can be properly filled and sealed. It is recommended that the main canal(s) and isthmus be prepared to a depth of 3 mm to enhance disinfection and the sealing by the filling material (Carr 1997). In the present case, the isthmus was apparently prepared only for a limited depth, which may help explain the surgical outcome observed.

In teeth with posttreatment apical periodontitis, the clinician is usually faced with the need to decide between retreatment, surgery or extraction. The CBCT scans revealed a missed distobuccal canal that could potentially be negotiated with the aid of the microscope. On the other hand, the mesiolingual canal was blocked and attempts to bypass it might have beeen unfruitful, with the risk of weakening the root structure. Surgery was indicated in the present case following the indication when retreatment is not feasible or when it is not expected to improve the results beyond what has already been achieved by the previous treatment (Gutmann & Harrison 1991).

In conclusion, the postsurgical apical periodontitis associated with the distal root was highly likely to have been caused/maintained by a residual infection, favoured by substandard technical issues. It was also quite apparent that the bioceramic material used as root-end filling may have lost its antimicrobial effects and sealing ability with time, allowing bacteria to leak across and grow on it.