Comparative evaluation of the accuracy of electronic apex locators and cone-beam computed tomography in detection of root canal perforation and working length during endodontic retreatment

Abstract

Background

To evaluate the accuracy of the electronic apex locators (EALs), and Cone-Beam Computed Tomography (CBCT) scanning, both in working length (WL) determination and in the detection of root canal perforations in retreatment cases.

Methods

Sixty human mandibular premolars were selected. After crown removal partially and canal access, root canals were instrumented and irrigated. The obturation process utilized gutta-percha and sealer with warm vertical compaction. Two groups were distinguished: one without perforation (Group 1) and the other with an apical third perforation (Group 2). Retreatment included filling removal, apical preparation, and irrigation. Actual working lengths (AWL) were determined using a stereomicroscope. CBCT images were used to measure CBCT working length (CWL), with adjustments for optimal views. Propex II and Dentaport ZX were used to measure electronic working length (EWL). Differences between EWL and AWL, as well as CWL, were analyzed to gauge accuracy. Data underwent Two-way ANOVA analysis. Measurements within ± 0.5 and ± 1 mm tolerance ranges were deemed successful for each device, followed by applying the Pearson Chi-square test.

Results

The study reveals no significant inter-group variations in device performance (p > .05). Dentaport ZX missed detecting perforation in two Group 2 (apical perforation) cases. For ± 1 mm tolerance, Propex II displayed the highest success in Group 2 (apical perforation).

Conclusion

This study demonstrates the comparable performance of Propex II, Dentaport ZX, and CBCT in endodontic retreatment, providing insights into diagnostic reliability.

Introduction

A root perforation that has a detrimental impact on the long-term success of root canal-treated teeth refers to an abnormal connection between the root canal system and the surrounding periodontal tissue [1]. This condition can arise due to various factors, such as internal or external root resorption, invasive dental decay, or accidental damage during root canal treatment or post-space preparation [2].

Accurate detection and precise localization of root perforations are paramount for ensuring a positive long-term prognosis. By precisely locating the perforation site, excessive instrumentation during root canal therapy can be avoided, reducing the possibility of irritating materials like irrigation solutions, sealers, or debris being pushed into the surrounding tissues [3]. Statistics suggest that the second most common cause of endodontic treatment failures is attributed to root canal perforations [4, 5]. If a perforation goes unnoticed and an infectious process ensues, the prognosis for successful treatment becomes precarious, and complications can become severe enough to necessitate tooth extraction. It underscores the importance of early detection and appropriate management to preserve the tooth’s health and function [6].

A precise determination of the working length (WL) is also crucial during the retreatment process, as it facilitates the complete removal of the primary root canal filling material to provide sufficient obturation [7]. Bergenholtz et al. [8] demonstrated that over-instrumentation and overfilling of retreated root canals significantly reduced the chances of achieving complete regeneration and repair of apical lesions in retreatment cases.

Electronic apex locators (EALs) are reliable tools for precise working length (WL) determination in endodontic treatments, including both initial root canal procedures and retreatments. They are also effective in localizing horizontal and oblique root fractures and detecting root canal perforations [8]. However, it’s important to note that the accuracy of EALs can be influenced by factors like debris, organic waste, calcium hydroxide, gutta-percha, sealer, and solvents [3, 9, 10]. Clinicians should be mindful of these factors to ensure optimal EAL performance in various clinical scenarios. Recently, medical professionals have proposed using Cone-beam computed tomography (CBCT) images already collected for prior operations to estimate the WL during root canal therapy [11]. Additionally, CBCT produced trustworthy results for diagnosing root perforation in previous investigations [12] CBCT makes it possible to view teeth and surrounding periodontal regions in various sections without distorting or superimposing anatomical components [13]. However, the shortcomings of CBCT images in teeth having endodontic therapy and the detection of root fractures represent the technology’s drawbacks.

To prevent further damage to the tooth and bone tissue, root perforations must be found as soon as possible [14]. While some studies have examined EALs’ accuracy in retreatment for WL determination [7, 14, 15] and perforation detection in untreated teeth [3, 16], none have specifically evaluated the accuracy of EALs and CBCT scanning in detecting root canal perforations during retreatment. This study aims to assess the accuracy of Dentaport ZX, Propex II, and CBCT scanning in determining WL and detecting root canal perforations in retreatment cases. The null hypothesis suggests no difference in accuracy among these devices for WL determination and perforation detection.

Materials and methods

The ethical approval of this study was made by The Clinical Research Ethics Committee of Akdeniz University Faculty of Medicine with the decision number KAEK-621. All participants provided informed consent to participate in the study. The sample size was calculated using G*Power 3.1.9.7 software (Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany) with a power of 0.85 and an alpha-type error of 0.05 using an effect size d of 0.25.

Selection and preparation of the teeth

Preoperative periapical radiographs were taken in both the mesiodistal and buccolingual directions to confirm the presence of a single and oval-shaped canal. Sixty mature human mandibular premolars with similar root canal width, straight root canals, and without caries, root canal resorption, calcification, fractures, or cracks were selected for the study. After the teeth were numbered from 1 to 60, randomization was achieved by adding 1 tooth to each group respectively. Mandibular premolars were selected because they often have a single root and canal structure. The crowns of all samples were partially removed under air and water cooling by using a diamond bur to obtain a flat reference plane and approximately standardized length of 17 ± 1 mm. After the endodontic access cavities were prepared, the WL was determined to be 1 mm shorter than the first time a size 10 K-file was seen apically under a stereomicroscope (Stemi 508, ZEISS, Germany). A size 10 and 15 K-file and WaveOne Gold Primary (0.25/07) (Dentsply Maillefer, Ballaigues, Switzerland) were used for instrumentation, respectively. Between each of these files, 2 mL of 2.5% NaOCl followed by 2 mL of sterile saline solution were used for irrigation. For the final irrigation, 2.5 mL of 17% EDTA during 1 min followed by 5 mL of sterile saline solution, and root canals were dried with paper points.

Obturation and preparation of perforation area

All the teeth were filled using a warm vertical compaction technique (Woodpecker Obturation System, Woodpecker Medical Instrument Co, Guilin, China). The apical portion was obturated with WaveOne Gold Primary (0.25/07) gutta-percha (Dentsply Maillefer, Ballaigues, Switzerland) and Dia-Proseal (Diadent, Cheongju, Korea) sealer which was introduced with the master cone. Fi-P Obturation Pen (Woodpecker Obturation System, Woodpecker Medical Instrument Co, Guilin, China) was used to cut gutta percha within 4–5 mm of WL, and the remaining root canal space was backfilled with Fi-G Obturation Gun (Woodpecker Obturation System, Woodpecker Medical Instrument Co, Guilin, China). The homogeneity of root canal filling was controlled with a periapical radiograph. Then, all the samples were stored at 100% humidity and 37°Cfor one week to provide the sealer to set.

Sixty teeth were divided into two groups according to the presence of the perforation area, randomly. The teeth in Group 1 had no perforation area and teeth in Group 2 (apical perforation) had a perforation area which was located at the apical third of the root (3 mm above the apex). To simulate the root canal perforation due to root canal resorption in teeth with root canal treatment, the diamond burs with a diameter of 1 mm were used to create the perforation area located at the lateral root surface. The diamond bur was advanced perpendicular to the root canal until the gutta-percha was seen in the canal space. To prevent differences that may occur due to abrasion of the bur, 1 new bur was used in 5 specimens.

Retreatment process

The root canal filling was removed using the D1 (0.30/0.09) at the cervical third, D2 (0.25/0.08) at the middle third, ProTaper Universal Retreatment files (Dentsply Maillefer, Baillagues, Switzerland) at 2-Ncm torque and 500-rpm speed. Finally, apical preparation was completed with WaveOne Gold Medium (0.35/0.06) and Large (0.45/0.05) files. Sterile saline was used as an irrigation solution during the retreatment process.

Determination of actual working length

In Group 1(no perforation) the actual working length (AWL) was recorded by measuring the distance between the rubber stop and the tip of a size 40 K-file when the file was first seen at the apical foramen with a digital caliper, under a stereomicroscope. In Group 2 (apical perforation), the AWL was determined by measuring the distance between the rubber stop and the tip of a size 55 K-file when the file was first seen at the perforation area.

Determination of electronic length

A freshly prepared alginate mold was used for electronic measurement. The teeth were placed in the mold until the enamel-cement junction and the lip clip of the Dentaport ZX (Morita Co, Japan) and Propex II (Dentsply Maillefer, Tulsa, OK, USA) were inserted in the mold, as well. During the electronic measurement, a size 40 K-file and a size 55 K-file, which were adapted adequately to apical foramen and at the perforation site, were used in Group 1 (no perforation) and Group 2 (apical perforation), respectively. All devices were calibrated before use by the manufacturer’s instructions and measurements were taken with the devices fully charged. The file was inserted through the canal of each tooth in the presence of sterile saline solution (0.9%) until a “0.0” signal was observed on the display of Dentaport ZX and the “Apex” signal was shown on the display of Propex II. These signals were accepted as apical foramen and perforation for Groups 1 and 2. When these signals were obtained, the rubber stop of the K-file was fixed to the coronal edge and the distance between the tip of the file and the rubber stop was measured after 5 s of stability. Thus, the electronic working length (EWL) was determined and recorded. Next, AWL was subtracted from the EWL, and the resulting values indicated whether the EWL was shorter or longer than the AWL, based on positive and negative values, respectively.

Determination of working length on the CBCT

To perform a CBCT scan, four teeth were inserted into the empty right and left premolar sockets of a dry human mandible that had modeling wax applied to simulate soft tissues. The VeraViev X800 (J. Morita, Kyoto, Japan) CBCT system was used to capture CBCT images of the teeth. Its settings included a tube voltage of 100 kV, a tube current of 4.8 mA, a voxel size of 0.080 mm3, a field of view (for) of 40 × 40 mm, and an exposure period of 17.86 s. For the CBCT measurements, One Volume Viewer software (J. Morita, Kyoto, Japan) was applied. Slices of teeth were examined. In Group 1 (no perforation), the teeth were adjusted in buccolingual directions for the optimum view. In Group 2 (apical perforation), the teeth were adjusted in mesiodistal and buccolingual directions to obtain the best view of the incisal edge, the entire canal, and the perforation area for Group 2 (apical perforation). WL was achieved by drawing a line from the incisal edge to the end of the root canal space for Group 1 (no perforation) and from the incisal edge to the beginning point of the perforation area for Group 2 (apical perforation) (Fig. 1). All the WL measurements were repeated twice by two experienced operators. The average value was calculated and noted for each tooth as CBCT working length (CWL). The computations were then performed by deducting AWL from CWL to calculate the differences.

Fig. 1.

A The WL measurement of a tooth in Group 1(no perforation) on mesiodistal section, B. The WL measurement of a tooth in Group 2 (apical perforation) on mesiodistal section, C. The WL measurement of a tooth in Group 2 (apical perforation) on vestibulooral section. The yellow line indicated the measurement from the incisal edge to the end of root canal space or the begining of perforation area

Statistical analysis

Two endodontists reviewed CBCT images and estimated the distance from the incisal edge to the end of the root canal space or the start of the perforation area twice. The inter-operator reliability for these measurements was high (Pearson correlation coefficient, 0.98). With operator Pearson correlation coefficients of 0.989 and 0.995 for operators 1 and 2, respectively, the intra-reliability was similarly strong.

The statistical analyses were made using SPSS 23.0 (SPSS Inc., Chicago, Illinois, USA). The Shapiro-Wilk normality test and Levene’s variance homogeneity test were applied to the obtained data. The data were normally distributed, with homogeneity of variance among groups. Data were analyzed using the Two-way ANOVA. The difference values within the ± 0.5 and ± 1 mm tolerance range were accepted as successful measurements for each device and the Pearson Chi-square test was applied.

Results

The two-way ANOVA results indicated that the presence of perforation is a significant factor (p = .010), while the measurement device is not (p = .087). Additionally, their interaction does not have a significant effect (p = .894).

In Table 1, the mean difference between the EWL/CWL and AWL of Group 1 (no perforation) and Group 2 (apical perforation) with the standard deviation (SD) for Propex II, Dentaport ZX, and CBCT were shown in detail. In Group 2 (apical perforation), Dentaport ZX could not detect the perforation area in two samples according to accepted criteria. There were no significant differences among the devices (p > .05).

Table 1.

The mean distances between the AWLs and the CWLs/EWLs were obtained by Propex II, Dentaportz ZX, and CBCT for Group 1 (WL: from the incisal edge to the apical foramen) and Group 2 (WL: from the incisal edge to the beginning point of the perforation area) during retreatment

DEVICES	Group 1 (no perforation)		Group 2 (apical perforation)		P*
	Mean	SD	Mean	SD
Propex II	-0.6342	± 0.5570	-0.3740	± 0.5602	0.894
Dentaport ZX	-0.57	± 0.6330	-0.3627	± 0.4639
CBCT	-0.7569	± 0.4617	-0.5909	± 0.4576

*P value indicated that the interaction between the devices used in this study and the presence of perforation was not significant

In Table 2, the success rate of each EAL and CBCT for ± 0.5 mm and ± 1 mm tolerance ranges were reported. There was no significant difference between the groups for all the devices (p > .05). The order of success was Propex II, Dentaport ZX, and CBCT for ± 1 mm tolerance range for Group 2 (apical perforation).

Table 2.

The success rate of CBCT, Propex II and Dentaportz ZX in the tolerance range of ± 0.5 and ± 1 mm for Group 1 (no perforation) and Group 2 (apical perforation)

DEVICES	± 0.5					± 1 mm
	Group 1		Group 2		p	Group 1		Group 2		p
	%	n	%	n		%	n	%	n
Propex II	40	12	30	9	0.417	73.3	22	86.7	26	0.197
Dentaport ZX	46.7	14	50	15	0.796	80	24	83.3	25	0.739
CBCT	30	9	33	10	0.781	70	21	76.6	23	0.559

Discussion

The physiological or pathological removal of cementum and/or dentine by tooth-resorbing cells is referred to as root resorption. Pathological root resorption can be brought on by several situations, including orthodontic treatment, trauma, replantation, late and erratically erupted teeth, or cysts and tumors. However, inflammation brought on by microbes is commonly linked to root resorption [17]. The migration of bacteria and their toxic byproducts from the root canal space to the periodontal ligament may also be facilitated by accessory canals, resulting in tissue inflammation [18]. In the advanced stages of this disease, root canal perforation due to resorption may occur. The accuracy of two distinct EALs and CBCT scans was assessed in the current study about both the detection of root canal perforations and the determination of working length in retreatment cases.

According to the study’s findings, there was no difference in the accuracy of the devices evaluated for determining the working length and perforation area, and the null hypothesis was accepted.

Nonsurgical endodontic retreatment is an important treatment choice for persistent periapical diseases. The complete removal of the root canal filling materials, disinfection, and obturation of the root canal system are important stages of this process for a long-term prognosis [18]. However, it was reported that removing all root canal filling materials could not be possible in previous studies [19, 20]. Furthermore, in a study conducted by Alves et al. [21], it was shown that residual materials were present in all the teeth and it was impossible to distinguish the tip of the file in the filling material on conventional radiographs during retreatment. In addition, according to some studies, the root canals that contain debris, irrigation solutions, and obturation materials can influence the accuracy of EALs [3, 22, 23]. There are some studies have evaluated the accuracy of EALs in determining the WL during retreatment [8, 15, 20, 23]. Still, there is no study that evaluates the accuracy of different EALs and CBCT scanning both in WL determination and in the detection of root canal perforations in retreatment cases.

To evaluate the accuracy of EALs under in vitro conditions, various electro-conductive materials such as saline solution, alginate mold, agar-agar, and gelatin were selected [14, 24–26]. In this present study, all the teeth were embedded in a freshly mixed alginate mold for electronic measurements to simulate both periodontal tissues and to prevent the operator can observing the roots during the experiment. In addition, to prevent the effect of gutta-percha solvents and irrigation solutions on the accuracy of EALs, sterile saline was used during the retreatment process and electronic measurement. In addition, to simulate the soft tissue and prevent imaging errors, the human mandible is covered with wax during CBCT imaging [13]. The diagnosis of root perforation is challenging to make clinically; however, the EALs are a technological tool that could be useful [27]. The effectiveness of EALs as a tool for detecting root perforation under various circumstances has been demonstrated in earlier investigations [3, 16, 26]. Nevertheless, no prior study has evaluated the accuracy of EALs in perforation detection during endodontic retreatment.

Studies evaluating the accuracy of various EALs and CBCT in the context of determining WL have been published in the literature. According to Üstün et al. [28], there was not a significant distinction between Propex-Pixi, Raypex 6, and CBCT in teeth with severe periapical lesions. In previous studies [29, 30], a strong correlation was also found between the measurements with Root ZX and CBCT scanning. Similar to these previous studies, there was no significant difference among the WL measurements for Propex II, Dentaport ZX, and CBCT imaging in Group 1 (no perforation).

The proper handling of CBCT images can show anomalies that are challenging to find in traditional periapical radiographs [31, 32]. The clinical findings should always be taken into consideration when making the final diagnosis and selecting the clinical therapies for these root perforations. The main advantage of CBCT exams is the potential to simultaneously observe all of the surfaces’ various planes and the position of each tooth [33]. An important factor to take into account is whether a tooth that has received endodontic therapy has a root perforation [2]. It may be challenging to identify root perforation in teeth that have undergone endodontic therapy. Diagnostic mistakes are a serious problem that frequently occurs. These mistakes usually co-occur with metallic or solid structures that are denser, leading to an image artifact that lacks homogeneity and can be identified by image contrasts [34]. In addition, CBCT imaging has certain drawbacks as well. The resolution of images is influenced by the voxel dimensions and the presence of artifacts caused by patient motion during scanning or the presence of radiopaque materials, such as root canal fillings, metal supports, and restorations [35]. Although there is a study [13] has shown the accuracy of CBCT in perforation detection in the literature, no previous study has been conducted to investigate the detection of perforations in retreated teeth.

According to the knowledge of the literature, there is no other study comparing the accuracy of EALS and CBCT in teeth with endodontic retreatment. In a prior study [14], the working length between the incisal edge and the root canal perforation area was calculated using the CBCT system and EALs. It was noticed that the perforation region with a diameter of 1 mm could be detected more successfully using CBCT than EALs. Although EALs were more successful than CBCT in terms of mean value and % success in the tolerance range of ± 1 mm, the difference was not statistically significant in the current study. This inconsistency may be due to the different CBCT devices and software, tooth selection, localization of the perforation area, or study design.

The accuracy of CBCT and EALs in the detection of perforations in teeth with endodontic retreatment has not been compared in the literature. As a result, the results of this study may offer crucial information for clinical studies that aim to diagnose root canal perforations during retreatment.

The inability to simulate the periodontal ligament, the EAL’s lack of contact with electrically conductive fluids like blood and saliva, patient movement during CBCT scans, and artifacts brought on by radiopaque materials like gutta-percha and root canal sealer as well as high-density nearby anatomical structures are some of the limitations of this in vitro study [16, 36]. Furthermore, despite the utilization of the human mandible and the application of wax to simulate the soft tissue, it fails to accurately replicate real-life conditions. In real-life scenarios, anatomical superpositions can have an impact on the photographs obtained, particularly in patient-related situations like patient movement during shooting. The diagnostic accuracy of CBCT and EALs in perforation identification in practical practice may be less than in the current study as a result of these limitations.

Conclusion

In conclusion, this present study compared the EALs and CBCT for endodontic retreatment cases. The results indicated similar performance among Propex II, Dentaport ZX, and CBCT devices in Group 1 (no perforation) and Group 2 (apical perforation). Dentaport ZX showed limitations in detecting perforation in Group 2 (apical perforation), but overall, there were no significant differences among the devices. Success rates within ± 0.5 mm and ± 1 mm tolerance ranges were consistent across devices, with Propex II, Dentaport ZX, and CBCT yielding similar results. This study adds to the understanding of EALs and CBCT in endodontic diagnostics, demonstrating their comparable performance and potential clinical utility.

Author contributions

HH and SK participated in designing the study. HH, SK, AK, participated in generating the data for the study. SK and HH participated in gathering the data for the study. SK and HH participated in the analysis of the data. SK and KH have had access to all the raw data of the study. SK, HH and AK have reviewed the pertinent raw data on which the results and conclusions of this study are based. SK, HH and AK have approved the final version of this paper. SK and HH guarantees that all individuals who meet the Journal’s authorship criteria are included as authors of this paper.

Funding

The authors report there are no competing interests to declare.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The Clinical Research Ethics Committee of Akdeniz University Faculty of Medicine with the decision number KAEK-621. All participants provided informed consent to participate in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.