Root canal morphology of the mandibular second premolar: a systematic review and meta-analysis

Background The aim of this paper was to systematically review the root canal configuration (RCC) and morphology literature of the mandibular second premolar (Mn2P). Methods Systematic research of five electronic databases was performed to identify published literature concerning the root canal configuration (RCC) of the Mn2P up through July 2020. Studies were selected according to predefined search terms and keywords inclusion criteria: “root canal configuration”, “root canal system”, “root canal morphology”, “mandibular second premolar”, “mandibular premolars”, “morphology” and “anatomy”. Further possible studies were identified by cross-referencing and screening the bibliographies of the selected articles. Results From 1622 retrieved studies, 44 studies investigating the internal morphology of 17,839 Mn2Ps were included. Most examined Mn2Ps were single-rooted (89.5–100%); two-rooted (0.1–8%) and three-rooted (0.1–3.5%) Mn2Ps at lower frequency. Most frequent RCCs reported were 1–1–1/1 (55.3–99.6%) followed by 1–1–2/2 (0.5–57%) and 2–2–2/2 (0.6–18%). The meta-analysis of seven studies demonstrated that a significantly higher number of RCC type 1–2–1/1 (OR [95%CI] = 2.05 [1.27, 3.33]) and 2–2–2/2 (OR [95%CI] = 2.32 [0.65, 8.63]) were observed in male than in female patients. Conclusions Different RCC research methods have been reported. Whereas clearing and radiographs were commonly used in the past, CBCT has been prevalent in recent years. A globally high frequency of a 1–1–1/1 RCC in the Mn2P has been reported. Nevertheless, the probability that different, more complicated RCCs can appear in Mn2Ps should not be underestimated and, thus, should be taken into consideration when making decisions during an endodontic treatment. Supplementary Information The online version contains supplementary material available at 10.1186/s12903-021-01668-z.

the classification described by Briseño Marroquín et al. [44]. However, a sizeable variation in the number of roots and root canals of the Mn2P was described in which the internal root canal morphology can be quite diversified [6,29,45].
A number of RCC investigations of the Mn2P have been carried out and analyzed with different research methodologies such as clearing [4,8,14,17,28,32,36,39], optical augmentation [32,44], cross-sectioning [5], radiography [5,25,30,38,41,43,46], CBCT [2, 3, 6, 9-12, 15-24, 26, 27, 29, 33, 34, 37, 42, 47] and dental computed tomography [41]. To the best of authors' knowledge, the root canal morphology of Mn2Ps by means of micro-computed tomography (micro-CT) has not been reported. Micro-CT has been described as a reproducible, non-destructive and non-invasive highresolution ex vivo method that, in association with 3D software imaging, is actually considered the most accurate root canal morphology research method [48] as well as the gold standard in endodontic internal morphology research [49]. The most frequently used root canal classification systems of Vertucci [40] and Weine et al. [50] are frequently reported; however, they are limited when describing an individual root canal morphology with precision, especially in cases of a complex root canal system. Therefore, a four-digit RCC system was created by Briseño Marroquín et al. [44]; the advantage of this RCC-system classification is that the classification system is a descriptive one and can be individually applied to the internal morphology of a specific tooth rather than forcing a classification based on the internal morphology system. The aim of this investigation was to undertake a systematic review of the literature concerning the root canal configuration of mandibular second premolars.

Methods
A systematic review to identify published literature concerning the root canal configuration (RCC) of the mandibular second premolar (Mn2P) until the end of July 2020 was carried out through a reference search of five electronic databases (Cochrane Database, Embase, MED-LINE/PubMed, Lilacs and Scopus) (Fig. 1). The current systematic review follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [51]. The review protocol was registered in the international prospective register of systematic reviews (PROSPERO) system (CRD42020192030, 14 July 2020).
Randomized controlled trials, cross-sectional, comparative, validation and evaluation studies of RCC's of Mn2Ps of different populations in patients of any age were included. Using a standardized comprehensive search strategy, the following Medical Subject Heading (MeSH) terms and keywords were used: "Root canal configuration" OR "root canal system" OR "root canal morphology" AND "mandibular second premolar" OR "mandibular premolars" AND "morphology" OR "anatomy". Additionally, other related studies were added by cross-referencing and hand searching the bibliographies of full text articles. The data collection was performed by an ad hoc-designed data extraction form without masking bibliographic record data, title, or authors. Only articles in English were considered. Studies in which teeth were only described as premolars or mandibular premolars without a clear assignment as well as case reports were excluded. After comparing the results from the five databases and the hand search, duplicates or repeated articles were rejected. Title and abstracts of the received articles were examined by two independent reviewers (A.L.A., T.G.W.) and if deemed relevant, the corresponding full text articles were consulted. Publication year and study duration, details/characteristics of the participants at baseline, and data regarding the RCC were recorded when available. The corresponding results, including relevant aspects, were summarized in tables. The obtained articles' abstracts, establishing whether the article should be excluded or included in the systematic review, were examined by two independent reviewers (A.L.A., T.G.W.). Thus, articles not matching the inclusion criteria were excluded. All articles meeting the inclusion criteria were retrieved in pdf format. The frequency of root canal configurations, the number of teeth, the number of roots, and the place of origin of the samples studied were presented in tables using the classifications of Vertucci [40], Weine et al. [50], and Briseño Marroquín et al. [44]. Briseño Marroquín et al. [44] RCC describe the root internal morphology in a coronal, middle and apical third direction by means of a four digits system. The first three digits are separated with a dash and represent the root canal number at the coronal boundary of the coronal, middle and apical third, respectively. The fourth digit is separated from the other three numbers with a slash and represents the number of apical main portals of exit. In addition, the different laboratory research methodologies that have been used by the different investigation groups were summarized in the tables as well.
The quality assessment of the included RCCs were assessed by two independent reviewers (A.L.A., T.G.W.) following the customized quality assessment tool developed by the National Heart, Lung, and Blood Institute (www. nhlbi. nih. gov/ health-topics/ study-quali ty-asses sment-tools). In case of disagreements between the independent reviewers, this has been discussed. If no consensus could be achieved, a third reviewer (R.J.W.) was consulted.
The risk of bias was assessed using the anatomical quality assessment (AQUA) tool for the quality assessment of anatomical studies included in meta-analyses and systematic reviews [52]. The same two authors (A.L.A., T.G.W.) screened the articles assessing the risk of bias; in case of disagreement in the assessment, the same author (R.J.W.) was consulted to reach consensus.
The Review Manager software (RevMan version 5.4 software, Cochrane Collaboration, Copenhagen, Denmark, 2014) was applied for the statistical analyses of the papers included into the meta-analyses. Odds ratio (OR) were chosen for calculating the effect size. The I 2 statistic was calculated to describe the percentage of variation across studies due to heterogeneity rather than chance [53]. Fixed or random-effects meta-analysis was performed depending on heterogeneity (I 2 < 35%: fixedeffects; I 2 > 35%: random-effect) [54,55]. The primary measures of effect between different root canal configurations, patient' sex and geographic reasons were Odds ratio and 95% confidence intervals (95% CI) for studies using dichotomous outcome data. Statistical significance was defined as a p value ≤ 0.05.

Results
The literature search of five different databases identified 1622 papers. After the results were compared and all duplicates were removed, 1255 articles were left in the initial search. Seventy-six studies that met the inclusion criteria were further considered after the title and abstract were consulted. After a full text analysis and adding articles retrieved by hand search, a total of 44 studies were included in this review (Fig. 1). These 44 morphology studies examined a total of 17,839 mandibular second premolars (Mn2Ps). The investigations included were carried out in different regions of the world and with different research methodologies.
The meta-analysis of seven studies sorted by geographical location by continent demonstrated that
Different root canal morphology research methods have also been compared. Khademi et al. [17] compared results from 182 mandibular first and second premolars with the clearing and CBCT techniques and reported that 87% of the results were in agreement with both research techniques. The highest agreement rate observed was in the 2-2-2/2 RCC (Vertucci's Type IV; Weine's Type III) and the lowest one in the 1-1-2/2 RCCs (Vertucci's Type V) groups. According to the authors, the CBCT technique demonstrated a higher accuracy than the clearing technique when recognizing C-shaped root canals but a lower accuracy in the recognition of lateral canals. Bolhari et al. [5] reported an agreement of 96.77 to 98.62% between bucco-lingual as well as mesio-distal projected radiographs and the cross-section technique. Regarding different ethnic groups, the comparative study Pedemonte et al. [29] reported that the 1-1-1/1 RCC (Vertucci's and Weine's Type I) was the most frequent one observed in Mn2Ps in Belgian (92.1%) and Chilean (95.0%) populations. Martins et al. [21] compared by means of CBCT the data obtained from a Chinese and a west European population and reported a slightly higher 1-1-1/1 RCC (Vertucci's and Weine's Type I) frequency in the Mn2Ps of the Chinese (99.6%) than in the west European (95.7%) groups. Using radiography, Trope et al. [57] investigated the RCC frequencies in 400 Mn2Ps in different ethnic groups and reported, at that time, that an Afro-American ethnic group (7.8%) had more than one canal more frequently than a Caucasian ethnic group (2.8%); however, these differences were not statistically significant. Yet, this study did not meet the inclusion criteria since it does not distinguish between different RCCs, and it was not included in the current systematic review. A comparative study [19] regarding different individual ages reported a slight 1-1-1/1 RCC (Vertucci's and Weine's Type I) decline from younger to older age groups (98.8% [21-40 years Although most Mn2Ps are single-rooted teeth, caution should always be exercised when attempting to compare the internal root canal morphology between different investigations since some authors do not report the number of roots observed. This precaution can be illustrated with Briseño Marroquín's et al. [44] 2-2-2/2 RCC, which describes the root canal morphology of one particular root, whereas Vertucci's [40] and Weine's et al. [50] classifications consider the tooth with its roots as a single entity.