Effect of quality of radiographs taken during root canal treatment on technical quality of root canal fillings and endodontic outcome
Article information
Abstract
Objectives
This study evaluated the number and quality of working length (WL) and master cone (MC) radiographs taken during root canal treatment by dental undergraduates, and their associations with the technical quality of root canal fillings (TQRCF) and endodontic outcomes (EO).
Methods
A retrospective evaluation of radiographs from 303 root canal-treated teeth in 231 patients was conducted, with 72 patients attending recall visits to assess EO. The chi-square and one-way analysis of variance tests were performed.
Results
A total of 505 WL and 557 MC radiographs were reviewed, with 72.9% and 75% deemed satisfactory, respectively. Satisfactory TQRCF was achieved in 60.4% of cases. Significant associations were found between the extension of the file in WL and gutta-percha in MC radiographs and TQRCF (p = 0.000). Misinterpretation of these radiographs resulted in poor TQRCF. Furthermore, 64.2% of teeth had satisfactory EO. A significant relationship was noted between the quality of MC radiographs and both TQRCF (p = 0.043) and EO (p = 0.003).
Conclusions
Unsatisfactory MC radiographs were linked to poor TQRCF and unfavorable EO. Regular radiographic training is recommended to enhance EO.
INTRODUCTION
Root canal treatment (RCT) is a key procedure in modern endodontics, focused on preserving natural teeth affected by pulpal and periapical diseases. Accurate determination of working length (WL) and appropriate selection of the master cone (MC) are essential steps for a successful RCT. To aid in WL determination, the European Society of Endodontology (ESE) recommends using electronic apex locators (EAL) for initial measurement, followed by a periapical radiograph to confirm the file’s position. The WL radiograph should record the complete area of interest, with the file tip positioned 0.5 to 1 mm short of the apex [1,2]. Additionally, the MC should extend to this length, and selecting a snugly fitting MC is critical for achieving optimal three-dimensional root canal obturation. This ensures a tight seal that prevents microbial infiltration and promotes periapical healing [3]. Research shows that ensuring the root canal filling reaches the ideal length is crucial for the long-term success of endodontic treatment. Fillings that extend to within 0 to 2 mm of the radiographic apex have significantly better prognoses compared to those that are either underextended or overextended [4,5]. Various other factors, including preoperative diagnosis, root canal anatomy, instrumentation techniques, quality of obturation, coronal seal, and operator skill, influence the technical quality of root canal fillings (TQRCF) and endodontic outcomes (EO) [6]. Furthermore, clinical and radiographic assessments are vital for successful EO, characterized by the absence of signs and symptoms such as pain, swelling, sinus tract, loss of function, and the presence of normal periodontal ligament space on radiographs [2].
The precise execution of these steps requires multiple radiographic assessments of high diagnostic quality for accurate interpretation and successful EO. However, increased radiographic assessments can lead to higher patient exposure to ionizing radiation and contribute to procedural inefficiencies. This concern may be balanced by adhering to the principle of “As Low as Diagnostically Achievable” [7,8].
Dental students often face challenges in obtaining periapical radiographs, particularly when establishing WL and selecting the MC, leading to the necessity for retakes. Common errors include inaccurate angulation, cone cuts, image distortion, inadequate exposure, overlapping anatomical landmarks, and rubber dam clamps. These issues can result in poor radiographic quality, which can lead to misinterpretations, procedural errors, and compromised EO [9,10].
These challenges underscore the importance of evaluating the knowledge, skills, and confidence of dental undergraduates (DUs) in performing oral radiology procedures. While several studies have examined these factors, results have varied significantly. For instance, Drage et al. [11] reported that dental graduates felt confident in their knowledge and skills, effectively applying radiological investigations in general dental practice. Conversely, Kavadella et al. [12] emphasized the need for reinforced training in oral radiology throughout the undergraduate curriculum, while Enabulele and Itimi [1] highlighted the limited knowledge and skills of dental students regarding endodontic radiology. This discrepancy suggests that, despite self-reported confidence, students may still struggle with practical aspects of radiographic techniques in endodontics, highlighting the critical need for further investigation into their radiographic practices.
Therefore, this study aimed to assess the number of WL (NWL) and MC (NMC) radiographs, as well as the quality of WL (QWL) and MC (QMC) radiographs taken during RCT procedures performed by DUs. Additionally, it sought to explore the associations between these radiographic parameters and the TQRCF and EO.
METHODS
Study design
Ethical approval for this retrospective-prospective study was obtained from the University’s Joint Committee on Research and Ethics (No. BDS I-01-2022[04]). Written informed consent was obtained from all patients as part of the root canal treatment process. Additionally, a separate written consent specifically for participation in the study was obtained from every patient. The retrospective element involved the collection of data from previous records from the dental management software (Open Dental Software, version 12.2; Open Dental Software, Inc., Salem, OR, USA) to identify patients who had undergone RCT, and then analyzing their radiographs to assess the NWL, NMC, QWL, QMC radiographs and TQRCF. On the other hand, the prospective component entailed recalling patients for a clinical and radiographic review of their root-treated teeth to evaluate the EO.
Population and sample size
The population for the retrospective part of the study involved the teeth of patients who received RCT from DUs at the Oral Health Centre (OHC), from January 2017 to December 2023. The sample size for the prospective part of the study (recalling patients) was calculated based on the mean and standard deviation using G*Power 3.1.9.2. With a 95% confidence interval and a 10% margin of error, the sample size was 68 participants.
Preclinical and clinical endodontic and radiology training
The endodontic curriculum at the university is delivered through lectures, case-based learning, seminars, and practical sessions. In their preclinical years, DUs are trained to perform RCT on extracted teeth and must pass a competency assessment before advancing to clinical practice at the OHC. In the clinical phase, they begin with single-rooted RCTs and progress to multirooted RCTs under endodontist supervision. RCT is carried out following the established protocol for the procedure [13].
Radiographic training for DUs begins in the second semester. They undergo training in intraoral radiographic techniques, including periapical and bitewing radiography. This comprehensive training involves didactic lectures, workshops, video demonstrations, and preclinical practical sessions on mannequins. During these sessions, students are guided through each step of the radiographic procedure: selecting the appropriate radiographic digital sensor and holder, inserting the sensor into the holder, positioning the holder with the sensor intraorally, immobilizing the patient to prevent image blurring, properly exposing and processing the sensor. Students continue to refine their imaging skills during clinical sessions from semesters 3 to 10.
To take an intraoral periapical (IOPA) radiograph using a paralleling technique, the dental student first ensures the patient is comfortably seated and informed about the procedure. They have the patient wear a lead apron and thyroid collar to minimize radiation exposure. The appropriate size digital sensor (phosphor storage plates) from Durr Dental (Durr Dental SE, Bietigheim-Bissingen, Germany) is selected and placed in a holder specifically designed for endodontic use (Kerr Endo-Bite Anterior and Kerr Endo-Bite Posterior holders; KerrHawe, Bioggio, Switzerland). It was then positioned in the patient’s mouth, centered over the area of interest, and aligned parallel to the long axis of the teeth. The X-ray tube is aligned so that the central ray is perpendicular to both the sensor and the tooth’s long axis, and a positioning device is used to maintain stability. The patient is instructed to remain still and bite gently on the holder while the X-ray machine is activated, following radiation safety protocols. After exposure, the sensor is processed and the digital image is retrieved for evaluation, ensuring clarity and diagnostic quality.
Root canal treatment procedure
All DUs were required to take a comprehensive history, perform necessary investigatory tests, and derive a diagnosis based on the findings. Subsequently, the findings were presented to a supervising endodontist for approval prior to commencing RCT. It was mandatory for students to perform RCT under rubber dam isolation. Preparation of access cavities was done using endodontic burs (Dentsply Maillefer, Ballaigues, Switzerland). WL determination was carried out using EAL (Root ZX II. J. Morita MFG. Corp., Kyoto, Japan) and reconfirmed with IOPA radiographs. Mechanical preparation of the root canal was done using manual K-files and ProTaper hand files (Dentsply Maillefer) followed by copious irrigation with 2% sodium hypochlorite, saline, and 17% ethylenediaminetetraacetic acid. Non-setting calcium hydroxide (PD Calcium Hydroxide; Produits Dentaires SA, Vevey, Switzerland) was placed as an intracanal medicament, and the access cavity was temporarily restored using a zinc oxide eugenol cement (Kalzinol; Dentsply, Konstanz, Germany) or reinforced zinc oxide eugenol (IRM; Dentsply Caulk, Milford, DE, USA) between appointments. During the final appointment, the canal was obturated with gutta-percha points and AH plus sealer (Dentsply Maillefer) using the cold lateral condensation technique. Thereafter, the coronal GP points were seared off with Touch N’ Heat (Kerr Dental, Brea, CA, USA) until the level of cementoenamel junction and compacted vertically. The most appropriate post-endodontic restorations (composite/post and core/crown) were then placed after determining the amount of remaining sound tooth structure.
Data collection
After obtaining the institutional ethical committee approval, all electronic records of RCT performed within this timeframe were retrieved from the electronic software records, Open Dental. Records of patients with incomplete RCT or those without a full set of digital periapical radiographs were excluded from the study.
Calibration training
Two examiners were trained by an endodontist and an oral radiologist to evaluate QWL, QMC radiographs, TQRCF, and EO prior to the initiation of the study. The examiners underwent a training and calibration exercise to ensure consistency and reliability in data collection. A pilot study was carried out on ten patients, and the examiners, endodontists, and oral radiologists independently carried out all observations. Each examiner’s reliability was tested using Cohen’s kappa test, and an inter-rater agreement kappa value of 0.945 was achieved. These examiners were blinded and did not have any access to any information about the clinician who performed the root canal procedure on these patients.
Participants recruitment
Electronic records of a total of 231 patients (303 teeth) were evaluated radiographically to assess NWL and NMC radiographs taken. Reasons for radiograph repetitions were documented. The quality of each WL and MC radiograph was evaluated based on seven key quality parameters (Table 1). Additionally, the pulpoperiapical status of each tooth prior to treatment was recorded to assess the initial condition of the pulp and periapical tissues before the intervention. Postoperative radiographs were used to assess TQRCF, focusing on extension, obturation homogeneity, taper, coronal seal quality, and the absence of procedural errors (Table 2). Technical quality was considered satisfactory if extension, homogeneity, taper, coronal seal quality, and procedural errors were within acceptable limits.
Based on the retrospective data, all patients who received RCT from DUs at the OHC, from January 2017 to December 2023 were contacted and scheduled for a follow-up assessment via phone calls, with up to three call attempts made for each patient. An explanation of the study’s purpose and the nature of the follow-up appointment was provided to the participants. The reasons for unwillingness to participate in the study were recorded. During the recall visit, clinical and radiographic evaluations were conducted to assess EO according to ESE guidelines (Table 3). Evaluation of all canals was done in multirooted teeth. The success of RCT was determined based on favorable outcomes observed clinically and radiographically. In cases where the EO was unsuccessful, these patients were referred to endodontists for further assessment and possible re-treatment if necessary.
Statistical analysis
Data was analyzed using IBM SPSS version 29 (IBM Corp., Armonk, NY, USA). Frequency distribution, chi-square, and one-way analysis of variance (ANOVA) tests were performed to assess associations between variables, with statistical significance defined as p < 0.05.
RESULTS
Demographic data of patients and teeth samples
The study analyzed records of 303 root canal-treated teeth from 231 patients, highlighting distinct demographic trends. RCTs were more common in females (57.1%) compared to males (42.9%). The majority of patients were over 46 years old (59.3%), followed by those aged 25 to 35 years (20.3%), 36 to 45 years (15.2%), and smaller proportions in the 19 to 24 years (2.6%) and under 18 years (2.6%) age groups. Maxillary teeth were predominantly treated (69.6%) compared to mandibular teeth (30.4). In terms of tooth type, premolars were the most treated (38.0%), followed by incisors (36.6%), molars (20.1%), and canines (5.3%). The most frequent pulpal diagnosis was pulpal necrosis (61.2%) followed by symptomatic irreversible pulpitis (25.0%), asymptomatic irreversible pulpitis (8.6%), and normal pulp in cases needing intentional RCT (5.2%). Asymptomatic apical periodontitis (60.1%) was the most common periapical diagnosis, followed by symptomatic apical periodontitis (27.4%), chronic apical abscess (8.9%), normal periapex (2.4%), and acute apical abscess (1.2%).
Among the 231 patients contacted, 72 patients (31.2%) participated in the follow-up assessment, during which 95 teeth were examined for their EO. The reasons for nonparticipation included failure to respond to phone calls, relocation to other states or countries, and various personal factors such as lack of interest, tight schedules, and transportation difficulties. Most participants were over 46 years old (63.9%), followed by those aged 25 to 35 years (16.7%), 36 to 45 years (12.5%), 19 to 24 years (4.2%), and under 18 years (2.8%). Premolars (38.9%) and maxillary teeth (66.3%) were the most frequently examined during recall visits. The follow-up duration for each participant ranged from 1 to 6 years, with an average of 2 years and 6 months.
Number of working length and master cone radiographs
For 303 root canal-treated teeth, a total of 505 WL radiographs and 557 MC radiographs were taken. This shows multiple WL and MC radiographs were taken to achieve accurate radiographs during the RCT procedure. The reasons for repeating radiographs were often related to underextension or overextension of WL or MC, the incomplete image captured, and the use of the buccal object rule. Ideally, one WL and one MC radiograph should be taken for one RCT procedure, at times, two attempts can be made to ensure the WL and MC. However, in the present study, it was evident that only 172 WL radiographs (56.8%) and 142 MC radiographs (46.9%) were satisfactory on the first attempt. In some cases, up to seven WL and six MC radiographs were taken for a single tooth RCT procedure. The number (frequency) of WL and MC radiographs taken for a single RCT procedure are illustrated in Table 4.
There was an underextension of the file in 119 teeth (23.5%) and overextension in 145 teeth (28.7%) in the first attempt of the WL radiographs. Similarly, an underextension of the gutta-percha cone in 131 teeth (23.7%) and overextension in 70 teeth (12.7%) were observed in the first attempt of the MC radiographs.
Quality of working length and master cone radiographs
The quality evaluation of WL and MC radiographs is detailed in Table 5. Of the 505 WL radiographs captured, 368 (72.9%) were deemed satisfactory. The most common issue leading to unsatisfactory WL radiographs was incomplete imaging of the area of interest, observed in 69 radiographs (13.7%). Among the 557 MC radiographs taken, 418 (75.0%) met the criteria for satisfactory quality. The primary challenge affecting MC radiographs was the superimposition of anatomical features and periapical pathologies, noted in 81 radiographs (14.5%).
Technical quality of root canal fillings
Overall, 183 teeth (60.4%) exhibited satisfactory TQRCF, while 120 teeth (39.6%) showed unsatisfactory TQRCF. Detailed results for each criterion used to assess TQRCF are presented in Figure 1.
Endodontic outcome
Out of 95 teeth of 72 patients assessed during the follow-up visit, 67 teeth (70.5%) showed satisfactory clinical outcomes, 81 teeth (85.3%) showed satisfactory radiographic outcomes, and 61 teeth (64.2%) showed an overall satisfactory EO. The detailed results for each criterion used in the clinical and radiographic outcome assessments are outlined in Table 3. TQRCF for these teeth (95 teeth) were again verified to assess the association with EO.
Associations between the number and quality of working length and master cone radiographs with the technical quality of root canal fillings and endodontic outcome
Cross tabulations were initially conducted for the number and quality of WL and MC radiographs against TQRCF, revealing no statistically significant association (p > 0.05) (Table 4). The subsequent analysis focused on assessing the relationship between WL and MC extension and TQRCF, identifying significant associations between file extension in WL and gutta-percha extension in MC radiographs with TQRCF (p < 0.001). Incorrect interpretation of WL and MC radiographs was implicated in contributing to poorer TQRCF outcomes.
For the patients in the follow-up assessments, cross tabulations for the number and quality of WL and the number of MC radiographs against TQRCF and EO showed no statistically significant associations (p > 0.05) (Table 6). However, a significant association was observed between the quality of MC radiographs with TQRCF (p = 0.043) and EO (p = 0.003). The unsatisfactory quality of MC radiographs was associated with poor TQRCF, resulting in unfavorable EO.
One-way ANOVA was performed to assess the association between seven parameters of QWL and QMC radiographs towards TQRCF and EO. There was no association found between the seven parameters of QWL radiograph with TQRCF and EO (Table 5). However, parameters such as the absence of superimposed anatomical features and periapical pathologies in the QMC radiograph showed significant association with the TQRCF (p < 0.001). Furthermore, there was a significant association between other parameters of QMC radiograph including a complete image of the area of interest seen (p = 0.020), absence of rubber dam clamp interference (p = 0.044), and appropriate quality (p = 0.044) with EO.
DISCUSSION
Precise establishment of WL is an essential factor in ensuring the success of RCT. Adequate shaping, cleaning, and sealing of the root canal system cannot be effectively achieved unless the WL is accurately determined. According to the American Association of Endodontists (AAE) guidelines, the MC should be adjusted to fit the length, apical size, and taper of the prepared root canal. An appropriately chosen MC should elicit a tug-back sensation, indicating resistance to displacement. Before proceeding with obturation, it is recommended to take a master apical cone radiograph of the master apical cone to confirm the length of the selected cone [14]. This study was designed to evaluate the number and quality of WL and MC radiographs and assess their impact on the TQRCF and the EO.
Approximately half of the sample successfully determined the WL and confirmed the appropriate cone fit with a single radiograph. However, a significant number of cases required multiple radiographs, with up to seven radiographs taken in some instances, to ensure accurate WL and cone fit. The primary reason for retakes was often overextension or underextension of the WL or MC. This could stem from complexities in root canal anatomy, inaccurate WL determination using EAL, or errors in recording the reference point. Employing a combination of electronic and radiographic methods has been shown to reduce the number of required radiographs and minimize patient radiation exposure in various studies [2,15]. Achieving precise EAL measurements relies on establishing a proper electrical circuit and understanding the electrical conduction properties of the canal [16]. It is crucial to remove any vital tissue, blood, or periapical exudate that could conduct electricity and cause measurement errors. For vital teeth, intact tissue should be removed, and excessive bleeding should be managed with paper points before using an EAL. For non-vital teeth with inflammatory exudate, it is recommended to remove the exudate using paper points or irrigate the canal with a solution containing negatively charged ions to prevent electrical interference [15,16] Furthermore, it is essential to establish straight-line access to the apex by performing coronal preflaring before taking WL measurements to ensure consistent WL [16]. Capturing radiographs for RCT can be particularly challenging due to the presence of rubber dam clamps and endodontic instruments, which may obstruct visibility and interfere with sensor or X-ray tube positioning. These challenges, compounded by students’ limited skill and experience, contribute significantly to the high rate of retakes for WL and MC radiographs [2].
According to the Conference of Radiation Control Program Directors guidelines, radiographic repeat rates ideally should be below 10% [17]. However, this study observed significantly higher rates with 40% for WL radiographs and 45.6% for MC radiographs, surpassing rates from similar studies. Yusof et al. [18] found a 34.3% repeat rate in routine radiographs among DUs while endodontic periapical radiographs showed even higher repetition rates, ranging from 35.6% to 51.9%, depending on the specific RCT step analyzed. In contrast, Nixon et al. [19] reported lower rates (5% to 11%) across various radiograph types, highlighting the variability in repeat rates across dental specialties. These findings emphasize the need for enhanced training in radiography skills, particularly in positioning techniques and the use of paralleling devices under rubber dam isolation, to improve accuracy and reduce repeat rates in endodontics.
The study reported satisfactory rates for the quality of WL (72.9%) and MC radiographs (75.0%). However, the incidence of specific parameters such as an incomplete image of the area of interest and anatomical superimposition was high, indicating a need for improved sensor and X-ray tube positioning. In WL radiographs, incomplete images of the area of interest were most common, due to cone cut, crown cut, and missed apices, consistent with prior research [20]. These errors typically result from improper positioning or patient movement during exposure. To prevent this, the X-ray beam should be directed perpendicular to the receptor, ensuring that the central ray and the area of interest align with the center of the receptor. For MC radiographs, the most frequent error noted was the superimposition of anatomical features, often due to bony structures such as sinus floor or zygomatic process overlapping the roots, hindering the visualization of periapical pathologies. Adjusting vertical angulation while using the bisecting angle technique or using the parallel cone technique can mitigate this problem. Another major error identified was the horizontal overlapping of roots, which stems from errors in horizontal angulation. Proper horizontal angulation, at zero degrees, is crucial for opening interproximal contacts and capturing a complete image. If the roots of multirooted teeth overlap despite using a zero-degree horizontal angulation, the buccal object rule can be employed to aid in the separation of the roots [20].
In this study, evaluation of TQRCF showed that 60.4% of treated teeth were technically acceptable. This is similar to other studies, which showed satisfactory root canal fillings in 42% to 55% of cases [21]. The primary reason for unsatisfactory root canal fillings was inadequate homogeneity, which aligns with previous research [22]. This could be due to the lack of sufficient lateral compaction of gutta-percha, inaccurate insertion of a spreader, and nonuniform coating of the root canal sealer over gutta-perch cones resulting in voids [23,24].
The current study demonstrated a statistically significant association between the extension of endodontic files in WL and gutta-percha in MC radiographs with TQRCF. Based on ESE quality guidelines, it is recommended to take another radiograph with readjusted file length in the canal if radiographic WL has been found to be short by more than 3 mm or overextended [2]. Yousuf et al. [25] highlighted the importance of maintaining accurate WL throughout RCT, as deviations can lead to overfilling and underfilling thereby affecting TQRCF. Chugal et al. [26] found the impact of deviation each millimeter from accurate WL increased the RCT failure rate by 14%. Similarly, MC should extend up to the WL, and MC periapical radiographs are recommended for precise obturation length determination to avoid both under and overextended root fillings. Sjogren et al. [5] showed that root fillings within 0 to 2 mm of the radiographic apex had a 94% normal periapical condition rate during follow-up. Conversely, fillings more than 2 mm short or beyond the apex showed decreased healing rates (68%–77.6%) [27,28].
In this study, satisfactory EO was achieved in 64.2% of RCTs by DUs. Tenderness to percussion or palpation was the most common clinical outcome, while soft tissue swelling, and sinus tract presence were least frequent. These results are similar to a study done in Brazil by da Rocha et al. [29] reporting a 60.7% success rate of RCT performed by dental students.
A statistically significant association was observed between QMC radiographs, TQRCF, and overall EO. The data revealed that a higher proportion of unsatisfactory MC radiographs correlated with poorer TQRCF and unfavorable EO, suggesting that QMC radiographs impacted the success of endodontic treatment. High-quality MC radiographs are vital for ensuring that the gutta-percha cone is positioned at the correct WL and properly adapted to the canal walls. This precise placement is essential for creating a tight apical seal, which helps prevent bacterial infiltration and lowers the risk of reinfection, reducing the likelihood of treatment failure and improving prognosis [3]. MC radiographs also enable clinicians to identify potential issues, such as underfilling or overextension, which can compromise the apical seal and increase the risk of complications like persistent infections or apical periodontitis [30,31]. The AAE underscores the importance of high-quality radiographs in treatment planning, as they provide critical diagnostic information that guides clinical decisions and enhances outcomes [32]. These findings highlight the essential role of MC radiographs in ensuring proper obturation, achieving satisfactory TQRCF, and securing successful EO.
Strength, limitations, and recommendations
This study’s correlation of the number and quality of WL and MC radiographs with TQRCF and EO represents a novel approach. However, a small sample size, due to patients’ unavailability for recall appointments, could affect the study’s generalizability. Additionally, the two-dimensional nature of periapical radiographs limited the assessment of EO, the use of cone-beam computed tomography to assess the EO would strengthen this research. By elucidating these relationships, this study provided insights into the importance of continuous reinforcement of oral radiology training in endodontics and clinical protocols to enhance the efficacy and reliability of endodontic treatments in dental education and practice.
CONCLUSIONS
A high number of WL and MC radiographs taken by DUs indicated a need for additional radiology training in the endodontic curriculum to reduce repetition rates and minimize patient radiation exposure. Achieving satisfactory WL and MC radiographs is crucial for satisfactory TQRCF. Poor quality of WL and MC radiographs correlates with poorer TQRCF and unfavorable EO, underscoring the importance of high-quality radiographs for successful outcomes.
Notes
Conflict of Interest
No potential conflict of interest relevant to this article was reported.
Funding/support
This project was funded by the International Medical University, Malaysia.
Acknowledgments
The authors would like to express our deepest appreciation to the Institute for Research, Development and Innovation (IRDI), International Medical University for funding this research. Furthermore, gratitude should also be expressed to the Oral Health Center, International Medical University for providing a location for us to carry out the research. Lastly, special thanks to all the patients who participated in the study.
Author Contributions
Conceptualization, Methodology, Project administration, Resources, Supervision: Parolia A, Chippagiri P. Data curation: Ng JM, Lee YY. Formal analysis: Ahanin E, Parolia A. Writing - original draft: Ng JM, Lee YY. Writing - review & editing: all authors. All authors read and approved the final manuscript.