This work attempted to determine which of two determinants (i.e., radiological quality of the root canal treatment or sealing of the coronal restoration) has a more positive effect on the periapical status. It associated a cross-sectional study and a literature systematic review.
Stemming from a much larger dataset, the sample size were limited, particularly those of subgroups, which would explain some unexpected results; e.g., a high rate of periapical lesions in teeth with good endodontic treatment and good coronal restoration might be simply due to the small sizes of group GE + GR (only 17 cases). Otherwise, group GE + PR had less periapical lesions than group PE + GR (1.1 times less) or PE + PR (1.9 times less). This indicated that a ‘diseased’ periapical status would be more frequent after inadequate root canal treatment than after inadequate coronal restoration (37.5% ‘diseased’ periapical status with GE vs. 42% with GR). The systematic review confirmed these results: group GE + PR had less periapical lesions than group PE + GR (1.7 times less) or group PE + PR (2.7 times less).
Though frequent in endodontics, cross-sectional studies seldom consider the kinetics of the periapical disease. Indeed, an X-ray revealed periapical lesion might be either in the process of development, in the process of healing, or a scar tissue (Kruse et al. 2017) (23). Besides, other clinical criteria might interfere with the interpretation of the data; e.g., initial pulp status, treatment procedure, time elapsed since treatment, time elapsed between the endodontic and the restorative procedure, etc.). Thus, cross-sectional studies might not answer all questions but help identifying potential risk factors for more rigorous clinical studies.
Here, a special attention was paid to two aspects: the representativeness of the patient sample and the relevance of the assessment criteria. One asset was that, unlike other same-design monocentric studies, the data stemmed from 13 hospital centers located in various French regions. According to Tiburcio-Machado et al. (2021), hospital data would yield same prevalence results as private practice data (51%; 95% CI: 40–63; I2 = 95.9% vs. 57%; 95% CI: 52–62%; I2 = 97.8%, respectively) (2). Anyway, the prevalence of apical periodontitis (all teeth, non-treated and root filled) in the general population remains high (40%; 95% CI: 33–46%; I2 = 96.5%) (2). Here, only root-filled teeth were considered and the prevalence of apical periodontitis was higher than 39% (95% CI: 36–43%; I2 = 98.5%) found by recent worldwide data (2).
Concerning quality criteria, those of root canal filling considered shape, density, and length(11) (AAE 2009). However, so far, most publications in systematic reviews have reported only on density and length. Here, considering each quality criterion alone, only inadequate density and inadequate length of root filling were found statistically significantly associated with periapical lesion whereas shape did not seem to significantly affect the periapical status. More investigations are needed to confirm or refute this finding.
Further, the present study assessed restoration quality on clinical and radiological criteria. This was found in only three studies kept by the systematic review (Dugas et al. 2003, Song et al. 2014, Hommez et al. 2002) (17, 19, 22). According to the FDI criteria, quality assessment of a coronal restoration should check for clinical or radiological evidence of hiatus between dental walls and the restoration material (9, 24), which reflects an inadequate sealing. The present investigation found 30.3% inadequate coronal restorations as per the radiological examination and 65.2% as per the clinical examination and both percentages are higher than those reported by Hommez et al. (2002) (21.8% and 24.8%, respectively) (17). This discrepancy indicates that radiological assessment of coronal restoration quality is not sufficient because of the risk of quality overestimation. A reliable assessment should use both assessment types. This confirms previous conclusions by Craveiro et al. (2015) (25). Indeed, the percentage of inadequate coronal restoration in this study (72.8%) was higher than those found by other studies that used only radiology (Moreno et al. 2013, Siqueira et al. 2005, Thampibul et al. 2019, Tavares et al. 2009) (16, 18, 20, 21). Still, the percentage of adequate restoration (27.2%) was lower than those found by Hommez et al. (2002) or Song et al. (2014) (67.4% and 68.5%, respectively) who used both assessment types (17, 19).
To assess periapical lesions, several recent epidemiological studies used Cone Beam Computed Tomography (CBCT) (Cakici et al. 2016; Gambarini et al. 2018; Gomes et al. 2015; Nur et al. 2014) (25–29) to overcome two conventional X-ray limitations: two-dimensional representation of three-dimensional structures (Bender 1997)(30) and confusion of periapical lesions with lesions confined to cancellous bone (Bender and Seltzer 2003 a and b) (31, 32). Despite CBCT superiority (Antony et al. 2020) (33), the current recommendations are not yet in favor of its systematic use (AAE and AAOMR Joint Position Statement 2015) (34) because of the ‘as low as reasonably achievable’ (ALARA) principle. Indeed, a small ‘field of view’ source exposes the patient to 10–100 times the radiation dose received with standard periapical radiography (35). Therefore, retroalveolar radiography remains the X-ray of choice, especially for epidemiological studies. In this study, the evaluation was as standard as possible (e.g., precluded X-rays without film holder) to avoid distorted images. Now, given that CBCT is not recommended, a more relevant approach to detect periapical lesions would be the use of artificial intelligence (36).
The present study found 27.3% adequate endodontic treatments, which is lower quality versus data from the systematic review; i.e., 33% (Moreno et al 2013) (16), 34.4% (Hommez et al. 2002) (17), 56.9% (Siqueira et al. 2005) (18), 50.5% (Tronstad et al 2000) (4), 35.6% (Song et al. 2014) (19), 51.9% (Thampibul et al. 2019) (20), 27% (Tavares et al. 2009) (21), and 38.9% (Dugas et al. 2003) (22). However, the present work’s data illustrate a slight increase in the percentage of adequate treatments from 21% (Boucher et al. 2002)(37) to 27% (Tavares et al. 2009)(21).
One important factor in quality is the level of experience of the practitioner. Craveiro et al. (2015) reported 82.9% of adequate root canal treatments by endodontists(25). This percentage is probably much lower in general dental practice. This highlights the need for collecting data from general practitioners to prevent overestimating the quality of endodontic treatments.
In this work, the literature review kept only eight studies. This low number is most probably due to the wide heterogeneity of the parameters evaluated, which is imposed by the multiple factors involved in endodontics. It is desirable that future prospective studies (in accordance with the ESE and FDI guidelines) be stratified on variables such as the type of tooth (indicative of the degree of technical difficulty), previous apical status, preoperative pulpal status, initial vs. secondary treatment, delay since last treatment, time elapsed before coronal seal, type of restoration (direct/indirect), or intra-operative factors. Such studies will highly clarify the importance of each variable in the quality of endodontic treatment.